These sound files accompany exercises, the treatment of intonation, and the description of some of the varieties of English given here. They are marked in the margins with a headphones symbol (as shown), and are available at: www.wiley.com/go/carrphonetics

gueqrecard_001

轨道 1.1：练习 4

轨道 1.2：练习 5

轨道 1.3：练习 6

Track 6.1：Marry Merry Mary（GA 中的元音中和）

轨道 10.3：示例 (3)（上升-下降）

轨道 10.4：示例 (4)（下降-上升）

轨道 10.5：示例 (5)（最后一个词项）

轨道 10.6：示例 (6)（最后一个词项）

轨道 10.7：示例 (7)（最后一个词项）

曲目 10.8：示例 (8)（对比语调）

轨道 10.9：示例 (9)（最后一个词项）

曲目 10.10：示例 (10)（对比语调）

曲目 10.11：示例 (11)（对比语调）

轨道 10.12：示例 (13)（已提供信息）

轨道 10.13：示例 (14)（已提供信息）

Track 10.14：示例（15）（同义词和给定信息）

曲目 10.15：示例 (16)（主音放置和预设）

曲目 10.16：示例 (17)（补品放置和给定信息）

Track 10.17: Example (18)（词尾时间状语）

Track 10.18: Example (19)（词尾时间状语和对比语调）

轨道 10.19：示例 (20)（前置时间状语）

Track 10.20：例子（21）（事件句）

轨道 10.21：示例 (22)（没有人，没有，没有地方，没有人）

Track 10.22: Example (23) ( someone, something, somewhere, someone )

轨道 10.23：示例 (24)（预科）

轨道 10.24：示例 (25)、(26)、(27)（分裂和聚焦）

轨道 10.25：示例 (28)（指示符）

轨道 10.26：示例 (29)（指示符）

轨道 10.27：示例 (30)（指示符和对比）

轨道 10.28：示例 (31)、(32)（非限制性和限制性关系从句）

Track 10.29: Example (33)（同位名词短语）

轨道 10.30：示例 (34)（其他括号）

轨道 10.31：示例 (35)（协调成分）

轨道 10.32：示例 (36)（短协调成分）

轨道 10.33：示例 (37)（词汇化协调）

轨道 10.34：示例 (38)（更多词汇化协调）

曲目 10.35：示例 (39)（列出语调）

轨道 10.36：示例 (40)、(41)（从句）

轨道 10.37：示例 (42)（句子状语）

轨道 10.38：示例 (43)（句子状语）

轨道 10.39：示例 (44)（伪裂缝）

Track 10.40: Example (45) (the is … is that construction)

轨道 10.41：示例 (46)（报告条款）

轨道 10.42：示例 (47)（报告条款）

曲目 10.43：示例 (48)（主语名词短语）

轨道 10.44：示例 (49)、(50)（标记问题）

曲目 10.45：示例 (51)（标记问题）

轨道 10.46：示例 (52)（标记问题）

曲目 10.47：示例 (53)（及物短语动词）

曲目 10.48：示例 (54)（不及物短语动词）

曲目 10.49：示例 (55)（不及物短语动词和事件句）

轨道 10.50：示例 (56)、(57)（程度状语）

Track 10.51: Example (58) （所以作为程度副词）

Track 10.52: Example (59) ( so as verb phrase adverbial)

Track 10.53：示例（60）（WH 问题）

轨道 10.54：示例 (61)（回声 WH 问题）

Track 10.55: Examples (62), (63)（陈述句作为问题）

曲目 10.56：示例 (64)（呼格）

曲目 10.57：示例 (65)（最终非呼格与最终呼格）

Track 10.58: Example (66) (IP boundaries and other meaning differences)

音轨 13.1：（送气、不送气、浊音和呼吸浊音）

音轨 13.2：（卷舌辅音）

轨道 13.3：（[v]、[υ] 和 [w]）

第一版序言

Prefaces to the First Edition

教师序

Preface for Teachers

每年在纽卡斯尔大学英语系，我都会得到 11 个 50 分钟的讲座时间，在第一年的第一学期向大约一百名学生介绍英语语音和音韵学，这些学生参加各种不同的本科学位课程，包括英语语言文学、语言学、英语语言、现代语言、音乐、历史等。学生团体中还包括欧洲交换本科生和攻读媒体技术应用语言学研究生学位和英语作为第二语言教师的语言学学位的学生。

Each year in the Department of English at Newcastle University, I am given eleven 50-minute lecture slots in which to introduce English phonetics and phonology to around a hundred students in the first semester of their first year on a variety of different undergraduate degree programmes, including English language and literature, linguistics, English language, modern languages, music, history and many others. Also included in the student body are European exchange undergraduates and students taking applied linguistics postgraduate degrees in media technology and in linguistics for teachers of English as a second language.

考虑到学位类型的范围，这是一项艰巨的任务，而且由于相当一部分学生的第一语言不是英语，这一任务变得更加困难。在典型的一年中，学生群体将包括讲阿拉伯语、法语、西班牙语、德语、希腊语、日语、韩语、普通话或粤语以及泰语的人。许多非母语人士将学习 RP；其他人将被教导一般美国人。在以英语为母语的学生中，很少有学生会说 RP，因此非母语者比母语者更可能说 RP。

Given the range of degree types, this is a daunting task, made even more difficult by the fact that a substantial minority of the students do not have English as their first language. In a typical year, the student cohort will include speakers of Arabic, French, Spanish, German, Greek, Japanese, Korean, Mandarin or Cantonese Chinese, and Thai. Many of the non-native speakers will have been taught RP; others will have been taught General American. Amongst the native speakers of English, very few of the students will be speakers of RP, so that the non-native speakers are more likely to speak RP than the native speakers.

绝大多数学生将不会进一步学习英语语音和音韵学，而这一多元化学生群体中的大多数人共有的一个因素是他们之前没有语音学或音韵学知识；因此，课程必须从头开始。

The vast majority of the student body will take their study of English phonetics and phonology no further, and the one factor which the majority of this diverse band of students shares is that they have no previous knowledge of phonetics or phonology; the course must therefore be ab initio.

在教授这样一门课程时，人们面临着两难的境地：一方面，人们想要迎合少数人的需求，他们将继续学习更高级的音韵学。另一方面，人们希望在介绍主题时不会用大量令人眼花缭乱的描述细节和大量看似神秘的理论结构来淹没学生。这一困境能否得到解决是一个有争议的问题。然而，这本教科书是作为一种解决方案的尝试而编写的。

One faces a dilemma in teaching such a course: on the one hand, one wants to cater to the small minority who will go on to study phonology at a more advanced level. On the other hand, one wants to introduce the subject without overwhelming the students with a mass of bewildering descriptive detail and an avalanche of seemingly arcane theoretical constructs. It is a moot point whether this dilemma can be resolved. However, this textbook was written as an attempt at a solution.

教科书比专着更难写是有争议的，而且越是初级的教科书越难写：勉强写下一行，却不知道自己经常有问题的假设，而且总是抵制诱惑在正文中质疑他们。一个人不断地感觉到一个人的同龄人在看着一个人的肩膀并且对一个人故意承诺印刷的荒谬的过度简化大笑。但又是必须要做的：学生要先学会走，才能学会跑；他们还必须先学会爬，然后才能学会走路。

It is arguable that textbooks are harder to write than monographs, and that the more elementary the textbook, the harder it is to write: one can barely write a line without being aware of one’s often questionable assumptions, and one has always to resist the temptation to question them in the body of the text. One continually has the sense of one’s peers looking over one’s shoulder and guffawing at the absurd oversimplifications which one is knowingly committing to print. But it has to be done: students have to learn to walk before they can learn to run; they also have to learn to crawl before they can learn to walk.

编写和使用教科书是一个经验问题：当练习、章节或书籍对特定的学生群体根本不起作用时，通常会立即显现出来。我在这里描述的一年级纽卡斯尔课程中使用的几乎所有教科书都以某种方式证明不适合这类学生群体；大多数情况下，它们包含了太多的细节。因此，我着手写一本非常简短、非常简单的教科书，它故意忽略了许多描述性/理论性的复杂性。

Writing and using textbooks is an empirical matter: it is very often immediately apparent when an exercise, chapter or book is simply not working, for a given body of students. Almost all of the textbooks which I have used on the first-year Newcastle course described here have proved to be unsuitable for this type of student cohort in one way or another; mostly, they have contained far too much detail. I have therefore set out to write a very short, very simple coursebook which deliberately ignores a great many descriptive/theoretical complexities.

我的目的不是向学生介绍音系理论；相反，我试图以一种尽可能与理论无关的方式介绍英语语音学和音系学的一些基本要素。当然，这从根本上是有问题的，因为不存在理论中立的描述。因此，我决定在纯实用主义的基础上采用各种理论/描述性观点，例如发音描述的舌弓/基元音方法、音段音位学的音位方法、英语脚结构的 trochaic 方法等我发现最容易传达给学生的内容。

My aim has not been to introduce students to phonological theory; rather, I have sought to introduce some of the bare essentials of English phonetics and phonology in a manner that is as theory-neutral as possible. This is fundamentally problematic, of course, since there is no such thing as theory-neutral description. I have therefore decided to adopt various theoretical/descriptive views, such as the tongue-arch/cardinal vowel approach to articulatory description, the phonemic approach to segmental phonology, the trochaic approach to English foot structure, and so on, on the purely pragmatic basis of what I have found to be easiest to convey to the students.

我忽略了声学语音学，原因很简单，因为我们系没有语音实验室，而且我没有包括显着特征，因为仅仅看到标有“+”和“-”符号的特征阵列似乎呈现了大量我的一年级学生头晕目眩（尤其是那些主修英语文学的学生）。我还排除了特征几何、拍子、规格不足和许多其他理论/描述性概念，试图将主题减少到最低限度。

I have ignored acoustic phonetics for the very simple reason that our department lacks a phonetics lab, and I have not included distinctive features, since the mere sight of arrays of features marked with ‘+’ and ‘−’ symbols seems to render large numbers of my first-year students dizzy (particularly those majoring in English literature). I have also excluded feature geometry, the mora, under-specification and a great many other theoretical/descriptive notions, in an attempt to pare the subject down to a bare minimum of these.

前四章确实有意写得很短，只包含对发音的最基本的介绍。我的目标是为学生提供对课程的温和介绍。我将音素原理的介绍分散在两章中，因为根据我的经验，学生发现他们第一次接触这些概念是一个巨大的飞跃。关于单词重音、节奏、连接的语音现象和口音变化的章节包含了对这些主题的非常精简、最少的说明；我希望那里有足够的内容作为那些希望更深入研究这些问题的学生的基础。在关于音节结构的章节中，我在介绍分析复杂性方面更加雄心勃勃，

The first four chapters are deliberately very short indeed, and contain only the most elementary introduction to articulatory phonetics. My aim there is to offer the student a gentle introduction to the course. I have spread the introduction of the phonemic principle over two chapters, since, in my experience, students find their first encounter with these ideas something of a quantum leap. The chapters on word stress, rhythm, connected speech phenomena and accent variation contain a very stripped-down, minimal, account of those subjects; I hope that there is enough there to act as a foundation for those students who wish to study these matters in more depth. In the chapter on syllable structure, I have been a little more ambitious in introducing analytical complexity, on the assumption that syllable structure is something that beginning students seem to be able to get the hang of more easily than, say, rhythm or intonation.

我认为大学教师最重要的职责之一是在学生中培养批判意识、对论证的把握和证据的作用。另一方面，在向初级阶段的学生介绍竞争分析的想法时必须非常谨慎：他们发现很难掌握一种分析的窍门，而不被要求评估优缺点竞争分析（即使在初级阶段，大多数本科生都非常反对批判性地比较不同分析的想法）。我试图通过在一两点引入相互竞争的分析和假设，同时在其他地方有意识地忽略它们来克服这种困境。

I believe that one of the most important duties of a university teacher is to induce in the student a sense of critical awareness, a grasp of argumentation and the role of evidence. On the other hand, one has to be very wary of introducing students at the most elementary stage to the idea of competing analyses: they find it difficult enough to get the hang of one sort of analysis, without being asked to assess the merits and demerits of competing analyses (even at the post-elementary stage, most undergraduates are very resistant to the idea of critically comparing different analyses). I have tried to overcome this dilemma by introducing competing analyses and assumptions at one or two points, while consciously ignoring them elsewhere.

这些练习将在每周的研讨会/辅导会议上进行讨论；我的经验是，如果不让语音学/音位学学生做练习，他们很容易相信他们已经掌握了这门学科，而事实上他们并没有。我希望完成本课程的学生能够找到更高级的教科书处理这些主题的方法，例如 Giegerich (1992) 和 Spencer (1996) 给出的方法。当然，这个希望是否实现在很大程度上是一个经验问题。

The exercises are meant to be discussed at weekly seminar/tutorial meetings; my experience is that, if phonetics/phonology students are not made to do exercises, they easily come to believe that they have grasped the subject when in fact they have not. It is my hope that students who have completed this course would find it possible to tackle more advanced textbook treatments of these topics, such as those given by Giegerich (1992) and Spencer (1996). Whether that hope is fulfilled is, of course, very much an empirical matter.

学生序

Preface for Students

这是对英语语音学和音韵学的初级介绍，专为那些以前对该主题一无所知的人而设计。它从发音语音学的非常基本的介绍开始，然后继续向学生介绍当今英语语音结构的一些主要方面的非常简单的说明。

This is an elementary introduction to English phonetics and phonology, designed for those who have no previous knowledge whatsoever of the subject. It begins with a very elementary introduction to articulatory phonetics, and then proceeds to introduce the student to a very simplified account of some of the main aspects of the phonological structure of present-day English.

有争议的是，人们在学习英语时可能会问两个主要问题：英语是什么使它成为一种语言（相对于非人类交流系统），以及英语是什么使它成为一种语言使它成为英语（而不是法语或韩语）？本书试图从英语的一个方面：它的音系学，为这两个问题提供答案的开端。

It is arguable that there are two main questions one might ask in studying the English language: what is it about English that makes it a language (as opposed to, say, a non-human communication system), and what is it about English that makes it English (as opposed to, say, French or Korean)? This book attempts to provide the beginnings of an answer to both of those questions, with respect to one aspect of English: its phonology.

因此，虽然这本书的主题是英语，但在几个地方也参考了其他语言的音韵学，通常是在对比练习中，这些练习旨在揭示英语和另一种语言之间的一个或多个差异。之所以包含这些对比练习，是因为以英语为母语的人通常对其他语言知之甚少或一无所知，他们倾向于认为英语的音韵是自然而然的，是必然的。对于许多这样的说话者来说，这似乎有些自然，例如，声音 [f] 相对于 [v] 的存在是为了表示含义上的差异（如fan与van). 因此，对于说英语的人来说，[f] 和 [v] 似乎很容易区分，而且这似乎也是一个自然事实。但是这些声音在英语中具有这种功能的事实是一个约定俗成的事实，而不是一个必然或自然的事实：英语不一定是那样的，而且可能并不总是这样那样吧正如一个人可以通过了解其他文化来获得对自己文化的新视角一样，一个人可以通过学习一点其他语言来获得对自己母语的新视角。在了解其他文化的过程中，人们还可以对人类文化有一些了解。同样，通过了解人类语言在哪些方面彼此相似，人们可以开始了解人类语言的音韵是什么样的。这些相似点涉及人类语言语音的一般组织特性，例如音位原则和音节结构原则。

Thus, although the subject matter of this book is English, there is reference to the phonology of other languages at several points, often in contrastive exercises which are designed to bring out one or more differences between English and another language. These contrastive exercises are included because native speakers of English, who often have little or no detailed knowledge of other languages, tend to assume that the phonology of English is the way it is as a matter of natural fact, a matter of necessity. For many such speakers, it will seem somehow natural, for instance, that the presence of the sound [f] as opposed to [v] functions to signal a difference in meaning (as in fan vs van). To the English speaker, [f] and [v] will therefore seem easily distinguishable, and that too will appear to be a natural fact. But the fact that these sounds have that function in English is a conventional, not a necessary or natural fact: English need not have been that way, and may not always be that way. Just as one can gain a new perspective on one’s own culture by learning about other cultures, so one can gain a fresh perspective on one’s native language by learning a little about other languages. One can also, in learning about other cultures, gain some sense of what human cultures are like. Similarly, one can begin to get a sense of what human language phonologies are like by learning in what respects they resemble each other. Those points of resemblance concern general organizational properties of human language phonologies, such as the phonemic principle and the principles of syllable structure.

阅读语言分析教科书不像阅读小说。在继续下一章之前，学生完成每章末尾的练习是至关重要的：它们旨在让学生应用本章介绍的思想。读者不会仅仅靠坐在扶手椅上阅读文本就正确地掌握了这本或任何其他音系学教科书所包含的思想，即使学生认为他们已经理解了这些思想。大量试图掌握语言分析但没有真正去做的学生最终以灾难性的考试结果告终：没有人在没有真正去做语言分析的情况下变得擅长语言分析。

Reading a textbook on linguistic analysis is not like reading a novel. It is vital that the student complete the exercises at the end of each chapter before proceeding to the next chapter: they are designed to get the student to apply the ideas introduced in the chapter. The reader will not have properly grasped the ideas contained in this, or any other, textbook on phonology by simply sitting back in an armchair and reading the text, even if the student is under the impression of having understood the ideas. Vast numbers of students who have attempted to master linguistic analysis without actually doing it have ended up with disastrous exam results: no one ever became any good at linguistic analysis without actually doing it.

与大多数语言学教科书一样，本书具有累积性：前几章介绍的内容是后几章的前提。因此，让几个星期过去而不做阅读和练习是致命的，希望以后能赶上来：结果很可能是你会发现自己力不从心，即使这是一本初级教科书。根本不可能像翻阅字典或百科全书那样翻阅语言分析教科书，无论它多么基础。

Like most linguistics textbooks, this book is cumulative in nature: what has been introduced in earlier chapters is presupposed in later chapters. It is fatal, therefore, to let several weeks go by without doing the reading and the exercises, in the hope of catching up later: the result is very likely to be that you will simply find yourself out of your depth, even though this is an elementary textbook. It is simply not possible to dip in and out of a linguistic analysis textbook, no matter how basic, in the way that one might dip in and out of a dictionary or an encyclopedia.

本书专为那些很可能不会继续学习英语语音和音韵学的学生而设计。然而，将进入更高级水平的学生应该能够处理这些主题的更高级的教科书处理方法，例如 Giegerich 和 Spencer 提供的处理方法（请参阅本书末尾的建议进一步阅读）。这些学生还应该发现更容易处理不以英语为重点的一般语音理论的许多介绍之一（再次，请参阅建议的进一步阅读）。为了让这些学生为更深入的学习做好准备，我在某些时候介绍了本书中某些假设的一些困难，或者对竞争分析的简短讨论。

This book is designed to cater for students who, in all probability, will not pursue their studies in English phonetics and phonology any further. However, students who will be proceeding to a more advanced level should be able to tackle more advanced textbook treatments of these topics, such as those given by Giegerich and by Spencer (see Suggested Further Reading at the end of the book). Those students should also find it easier to tackle one of the many introductions to general phonological theory which are not focused on English (again, see Suggested Further Reading). In order to prepare such students for more advanced study, I have introduced, at some points, an indication of some of the difficulties with some of the assumptions made in this textbook, or a brief discussion of competing analyses. Although this textbook merely scratches the surface of the subject matter, I hope that there is enough here to make the subject of phonology seem intriguing to the student who intends to pursue his or her studies.

我希望这本书对英语作为外语的教师有所帮助，尽管它不是专门为这些读者设计的。我总是很惊讶地发现这些老师通常对英语语音和音韵学的了解是如此之少。我没有这样的教学经验，虽然我没有就如何将本书中介绍的概念应用于 TEFL 课堂提出任何建议，但我发现很难相信一个知识英语语音学和音韵学的基础知识在某种程度上可能对 TEFL 教师没有用处，即使只是作为扩展教师英语知识的背景知识。我也希望一些对比练习可能有助于提出一些方法，说明一个人的母语音韵会干扰一个人将英语作为第二语言的尝试。

It is my hope that this book will be of some use to teachers of English as a foreign language, although it is not designed specifically for such readers. I am always surprised to discover how little in the way of knowledge of English phonetics and phonology such teachers often have. I have no experience of such teaching, and while I make no suggestions as to how the notions introduced in this book might be put to use in the TEFL classroom, I find it hard to believe that a knowledge of the basics of English phonetics and phonology could fail to be useful to the TEFL teacher in some way, even if only as background knowledge which extends the teacher’s knowledge of English. I also hope that some of the contrastive exercises might help suggest ways in which one’s native language phonology can interfere with one’s attempt to acquire English as a second language.

纽卡斯尔，1999 年 2 月

Newcastle, February 1999

第二版前言

Preface to the Second Edition

这本书的第一版是我在英国大学教书时写的。从那以后，我搬到了法国蒙彼利埃大学的英语系。虽然我在纽卡斯尔大学的课堂上总是有非英语母语的人，但我的大多数学生都是以英语为母语的；现在，我的绝大多数学生都不是以英语为母语的人。大多数是法语学生，但也有西班牙语、葡萄牙语、希腊语、德语、荷兰语、波兰语、俄罗斯语和保加利亚语的学生，等等。这本书因此发生了变化：它更面向将英语作为外语的学习者，但我相信它对以英语为母语的人仍然有用。

The first edition of this book was written while I was teaching in an English university. Since then, I have moved to the English department at Montpellier University, in France. While I always had non-native speakers of English in my classes at Newcastle University, most of my students were native speakers of English; now, the vast majority of my students are not native speakers of English. Most are French, but there are also Spanish, Portuguese, Greek, German, Dutch, Polish, Russian and Bulgarian students, among others. The book has changed as a result: it is more orientated towards learners of English as a foreign language, but it is still useful for native speakers, I believe.

文本的主要变化涉及后面的章节：第 8、9 和 10 章已完全重写，并且有一个新的章节（第 11 章）是关于拼写和发音之间的关系，称为 grapho-phonemics。如果学生的母语是英语，教师可以选择跳过本章，但对于希望继续教授英语作为外语的学生来说，这可能很有用。我扩展了附录（重命名为第 13 章）以涵盖其他英语种类。现在有伴随练习、语调处理和此处给出的一些英语变体的描述的声音文件：这些在页边空白处标有耳机符号。

The main changes to the text concern the later chapters: chapters 8, 9 and 10 have been entirely rewritten, and there is a new chapter (chapter 11) on the relationship between spelling and pronunciation, known as grapho-phonemics. Teachers whose students are native speakers of English may choose to skip this chapter, but it could prove useful for students who wish to go on to teach English as a foreign language. I have expanded the appendix (renamed as chapter 13) to cover additional varieties of English. There are now sound files which accompany exercises, the treatment of intonation, and the description of some of the varieties of English given here: these are marked in the margins with a headphones symbol.

gueqrecard_002

我坚持保留音标练习，原因有二。首先，我认为它加强了基于听语音的语音转录与音位分析之间的区别，在语音分析中，音素（如这里所设想的那样）不是语音，也不能被听到。其次，我希望本书的一些读者能够继续从事各种英语的实证研究，这通常包括仔细聆听和语音转录不同口音说话者的录音，同时也参与到理论问题中对这些口音的分析。音标练习现在以录音为基础。

I have insisted on retaining practice at phonetic transcription, for two reasons. Firstly, I believe that it reinforces the distinction between phonetic transcription, based on listening to speech sounds, and phonological analysis, in which phonemes (as conceived of here) are not speech sounds, and cannot be heard. Secondly, I hope that some readers of this book will go on to engage in the empirical study of varieties of English, which typically involves both listening carefully to, and phonetically transcribing, recordings of speakers of various accents, and also engaging with theoretical issues in the analysis of those accents. The phonetic transcription exercises are now based on audio recordings.

这本书的目的不是介绍音系理论；一些这类书籍列在“建议的进一步阅读”中。不可避免地，我不得不借鉴各种理论框架中提出的概念。任何提出的理论和描述之间的区别都充满了困难：正如科学哲学家卡尔波普尔所指出的那样，没有理论假设就没有描述。然而，在我看来，必须坚持理论和描述之间的某种区别。我的目标主要是描述性的。

The book is not intended as an introduction to phonological theory; some books of that sort are listed in the Suggested Further Reading. Inevitably, I have had to draw on notions proposed in various theoretical frameworks. Any proposed distinction between theory and description is fraught with difficulties: there can be no description without theoretical assumptions, as the philosopher of science Karl Popper pointed out. However, in my view, some kind of distinction between theory and description must be upheld. My aims here are primarily descriptive.

任何疑问和/或更正可以发送至：philip.carr@univ-montp3.fr

Any queries and/or corrections can be sent to: philip.carr@univ-montp3.fr

蒙彼利埃，2011 年 12 月

Montpellier, December 2011

致谢

Acknowledgements

致谢，第一版

Acknowledgements, First Edition

我很感谢纽卡斯尔大学英语文学和语言学系的几批学生，他们的反馈很有价值。我还要感谢 Patrick Honeybone、Maria Maza、Irenie Rowley 和 Charles Prescott，他们是我学生的导师，他们对几份草稿的评论被证明是最有帮助的。还要感谢我的同事 Karen Corrigan，她对早期草稿提出了意见。许多年前，在白尼罗河上一个岛上的一间小泥屋里，我向詹姆斯·狄金斯介绍了语音描述的元素。作为回报，他向我提供了第 1 页上的阿拉伯语数据。40，为此我感谢他。我还要感谢七位匿名审阅我的原始提案的评论，以及史蒂夫史密斯，感谢 Blackwell 的 Mary Riso 和 Beth Remmes 的耐心和鼓励。他们可能会惊讶于写这么短的书花了这么长时间；然而，我可以安慰自己，因为我不必提供（也许是伪造的）道歉：'我很抱歉这本书太长了；我没有时间写一个更短的。最后，非常感谢 Blackwell 的读者 Andrew Spencer，他对预定稿的睿智、见多识广和富有洞察力的评论非常有帮助。当然，他不对文本中任何遗留的不足之处负责。最后，非常感谢 Blackwell 的读者 Andrew Spencer，他对预定稿的睿智、见多识广和富有洞察力的评论非常有帮助。当然，他不对文本中任何遗留的不足之处负责。最后，非常感谢 Blackwell 的读者 Andrew Spencer，他对预定稿的睿智、见多识广和富有洞察力的评论非常有帮助。当然，他不对文本中任何遗留的不足之处负责。

I am grateful to several cohorts of students in the Department of English Literary and Linguistic Studies at Newcastle University whose feedback has been valuable. I am also indebted to Patrick Honeybone, Maria Maza, Irenie Rowley and Charles Prescott, who acted as tutors to my students, and whose comments on several drafts have proved most helpful. Thanks too to my colleague Karen Corrigan, who commented on an early draft. Many years ago, in a small mud house on an island on the White Nile, I introduced James Dickins to the elements of phonetic description. He has kindly reciprocated by supplying me with the Arabic data on p. 40, for which I thank him. I also owe a debt of gratitude to the seven anonymous reviewers of my original proposal for their comments, and to Steve Smith, Mary Riso and Beth Remmes at Blackwell for their patience and encouragement. They may well be surprised that it took so long to write such a short book; I can, however, console myself with the fact that I do not have to offer the (perhaps apocryphal) apology: ‘I am sorry this book is so long; I did not have the time to write a shorter one.’ Finally, many thanks to Blackwell’s reader, Andrew Spencer, whose intelligent, informed and insightful comments on the pre-final draft were immensely helpful. He is not, of course, responsible for any remaining inadequacies in the text.

致谢，第二版

Acknowledgements, Second Edition

我要感谢 PAC 项目 (La Phonologie de l'anglais contemporain) 的两位联合主管 Jacques Durand 和 Anne Przewozny，并感谢 Raphaël Dommange 允许他们使用他们的一些 PAC 录音。

I am grateful to Jacques Durand and Anne Przewozny, fellow co-directors of the PAC project (La Phonologie de l’anglais contemporain), and to Raphaël Dommange for permission to use some of their PAC recordings.

感谢我以前的硕士生 Cécile Montforte 允许我使用她论文“口音变化和语言态度”中的一些 PAC 数据在苏格兰：对格拉斯哥中产阶级演讲者的社会语言调查（蒙彼利埃第二大学，2011 年）。

Thanks to my ex-Masters student Cécile Montforte for allowing me to use some of her PAC data from her dissertation ‘Accent change and language attitudes in Scotland: a socio-linguistic investigation of Glasgow middle-class speakers’ (Université Montpellier II, 2011).

感谢剑桥大学语音学教授兼 IViE 项目（英语语调变化）首席研究员 Francis Nolan 允许我出于教育目的使用来自 IViE 语料库的两个声音文件，这些文件是作为 ESRC 的一部分收集的项目 R000237145“不列颠群岛的英语语调”。

I am grateful to Francis Nolan, professor of phonetics at Cambridge University and principal investigator of the IViE project (Intonational Variation in English), for permission to use, for educational purposes, two sound files from the IViE corpus, collected as part of the ESRC project R000237145 ‘English Intonation in the British Isles’.

非常感谢我在蒙彼利埃保罗瓦莱里大学的朋友和同事尼克迈尔斯，感谢伊内斯布鲁拉德和我的硕士生艾莉森吉尔布雷思，感谢他们花时间和麻烦为我记录练习材料。

Many thanks to my friend and colleague Nick Myers, of the Université Paul Valéry here in Montpellier, to Inès Brulard and to my Masters student Alison Gilbreath for taking the time and trouble to record exercise material for me.

感谢章节草稿的两位匿名读者提供的详细评论。

I thank two anonymous readers of the draft chapters for their detailed comments.

感谢 Wiley-Blackwell 的 Danielle Descoteaux 和 Julia Kirk 的耐心和鼓励。我还要感谢科琳·菲茨杰拉德 (Colleen Fitzgerald) 允许使用她为德克萨斯英语录音制作的书面段落，感谢我的文字编辑菲奥娜·塞维尔 (Fiona Sewell) 帮助我草拟文本，感谢莉亚·莫林 (Leah Morin) 帮助校对。

I thank Danielle Descoteaux and Julia Kirk at Wiley-Blackwell for their patience and encouragement. I also thank Colleen Fitzgerald for permission to use the written passage which she constructed for the recordings of Texan English, my copy-editor Fiona Sewell for her help with the draft text and Leah Morin for her help with the proofs.

图 1发音器官

Figure 1 The organs of speech

图2国际音标（塞萨洛尼基亚里士多德大学英语学院理论与应用语言学系，塞萨洛尼基54124，希腊）

Figure 2 The International Phonetic Alphabet (Department of Theoretical and Applied Linguistics, School of English, Aristotle University of Thessaloniki, Thessaloniki 54124, Greece)

1个

1 英语语音：辅音（一）

English Phonetics: Consonants (i)

1.1 气流和清晰度

1.1 Airstream and Articulation

语音是通过改变气流产生的。我们将在本书中关注的气流涉及空气从肺部通过口腔和鼻腔排出的通道（见图 1）。空气流可以在许多点上进行修改，并且可以通过多种方式进行修改（即以某种方式收缩）。当气流从肺部通过时，第一个可以改变气流的点是喉部（你可以感觉到喉结的前部，喉结，在喉咙前部略微突出；见图 1） , 其中位于声带(或声带). 声带可能是张开的，在这种情况下，气流可以畅通无阻地通过它们。从上面看，声带张开时是这样的：

Speech sounds are made by modifying an airstream. The airstream we will be concerned with in this book involves the passage of air from the lungs out through the oral and nasal cavities (see figure 1). There are many points at which that stream of air can be modified, and several ways in which it can be modified (i.e. constricted in some way). The first point at which the flow of air can be modified, as it passes from the lungs, is in the larynx (you can feel the front of this, the Adam’s apple, protruding slightly at the front of your throat; see figure 1), in which are located the vocal folds (or vocal cords). The vocal folds may lie open, in which case the airstream passes through them unimpeded. Viewed from above, the vocal folds, when they lie open, look like this:

打开声带

Open vocal folds

声带可以靠在一起，这样它们就关闭了，没有空气可以从肺部流过它们：

The vocal folds may be brought together so that they are closed, and no air may flow through them from the lungs:

闭合声带

Closed vocal folds

可以改变流出气流的一种方法是施加一定水平的恒定肌肉压力，足以使声带沿其长度闭合，但仅此而已；鉴于肌肉压力的程度，该闭合下方的气压积聚足以迫使该闭合打开，但气压随后下降，肌肉压力导致褶皱再次闭合。然后非常快速地重复该序列，并导致所谓的声带振动。如果将手指放在喉咙上并发出单词hazy中写为 <z> 的声音，您应该能够感觉到这种振动（尽管您可能也会在头部的其他地方感觉到振动）。据说这种声带振动产生的声音是浊音，而没有这种振动产生的声音被称为清音。

One way in which the outgoing stream of air may be modified is by applying a certain level of constant muscular pressure sufficient to close the vocal folds along their length, but only just; the build-up of air pressure underneath this closure is sufficient, given the degree of muscular pressure, to force that closure open, but the air pressure then drops, and the muscular pressure causes the folds to close again. The sequence is then repeated, very rapidly, and results in what is called vocal fold vibration. You should be able to feel this vibration if you put your fingers to your larynx and produce the sound which is written as in the word hazy (although you will probably also feel vibration elsewhere in your head). Sounds which are produced with this vocal fold vibration are said to be voiced sounds, whereas sounds produced without such vibration are said to be voiceless.

为了转录语音，语音学家使用国际音标（IPA：见图 2）；以朦胧形式写成 <z> 的 IPA 符号是 [z]。如果您将手指放在喉咙上并发出 [z]，然后是 [s]（如miss中），然后再次 [z]，您应该能够感觉到 [z] 中存在振动：[z] 是浊音，而 [s] 是清音。这种区别将构成三个描述性参数中的第一个，我们将通过它来描述给定的辅音语音：我们将说，对于任何给定的辅音，它是浊音还是清音。

To transcribe speech sounds, phoneticians use the International Phonetic Alphabet (the IPA: see figure 2); the IPA symbol for the sound written in hazy is [z]. You should be able to feel the presence of vibration in [z] if you put your fingers to your larynx and produce [z], then [s] (as in miss), then [z] again: [z] is voiced, whereas [s] is voiceless. This distinction will constitute the first of three descriptive parameters by means of which we will describe a given consonantal speech sound: we will say, for any given consonant, whether it is voiced or voiceless.

1.2 发音地点

1.2 Place of Articulation

我们将把气流可以改变的点称为发音部位。我们刚刚将声带确定为发音的地方；由于声带之间的空间被称为声门，我们将在这个发音部位产生的声音称为声门声音。还有许多其他的发音地方；我们将确定另外七个。

We will refer to the points at which the flow of air can be modified as places of articulation. We have just identified the vocal folds as a place of articulation; since the space between the vocal cords is referred to as the glottis, we will refer to sounds produced at this place of articulation as glottal sounds. There are many other places of articulation; we will identify a further seven.

首先，通过在下唇和上唇之间形成收缩来改变气流的声音被称为双唇音。一个例子是pit中的第一个声音。

Firstly, sounds in which the airflow is modified by forming a constriction between the lower lip and the upper lip are referred to as bilabial sounds. An example is the first sound in pit.

双唇音：入坑第一音

A bilabial sound: the first sound in pit

其次，下唇和上牙之间存在收缩的声音被称为唇齿音。一个例子是fit中的第一个声音。

Secondly, sounds in which there is a constriction between the lower lip and the upper teeth are referred to as labio-dental sounds. An example is the first sound in fit.

唇齿音：第一个合适的声音

A labio-dental sound: the first sound in fit

第三，舌尖和上牙之间有收缩的声音被称为牙音。一个例子是thin中的第一个声音。

Thirdly, sounds in which there is a constriction between the tip of the tongue and the upper teeth are referred to as dental sounds. An example is the first sound in thin.

牙音：薄中的第一音

A dental sound: the first sound in thin

对于其余发音部位，让我们区分舌尖、舌叶、舌前和舌后（如图 1 所示）。我们还要区分嘴巴上部的各个点。我们将确定四个不同的区域：牙槽嵴（牙齿后面坚硬的骨质脊；见图 1）、硬腭（口腔顶部坚硬的骨质部分；见图1）、腭-牙槽骨（或后牙槽）区域¹（牙槽嵴和硬腭之间的区域）和软腭（口腔顶部后部的柔软部分，也称为软腭；见图 1）。

For the remaining places of articulation, let us distinguish between the tip, the blade of the tongue, the front of the tongue and the back of the tongue (as in figure 1). Let us also distinguish various points along the upper part of the mouth. We will identify four different areas: the alveolar ridge (the hard, bony ridge behind the teeth; see figure 1), the hard palate (the hard, bony part of the roof of the mouth; see figure 1), the palato-alveolar (or post-alveolar) region1 (the area in between the alveolar ridge and the hard palate), and the velum (the soft part at the back of the roof of the mouth, also known as the soft palate; see figure 1).

舌叶或舌尖与牙槽嵴之间存在收缩的声音称为牙槽音。一个例子是sin中的第一个声音。

Sounds in which there is a constriction between the blade or tip of the tongue and the alveolar ridge are called alveolar sounds. An example is the first sound in sin.

肺泡音：罪恶中的第一个声音

An alveolar sound: the first sound in sin

舌叶和腭-牙槽（或后牙槽）区域之间存在收缩的声音称为腭-牙槽音。一个例子是ship中的第一个声音。

Sounds in which there is a constriction between the blade of the tongue and the palato-alveolar (or post-alveolar) region are called palato-alveolar sounds. An example is the first sound in ship.

腭音：船上的第一个声音

A palato-alveolar sound: the first sound in ship

舌头前部和硬腭之间存在收缩的声音称为腭音。一个例子是yes中的第一个声音（尽管这对您来说可能不太明显；我们将在下面返回到这个声音）。

Sounds in which there is a constriction between the front of the tongue and the hard palate are called palatal sounds. An example is the first sound in yes (although this may be less obvious to you; we will return to this sound below).

上颚音：yes中的第一个音

A palatal sound: the first sound in yes

舌后部和软腭之间有收缩的声音称为软腭音。一个例子是cool中的第一个声音。

Sounds in which there is a constriction between the back of the tongue and the velum are called velar sounds. An example is the first sound in cool.

软腭音：冷音中的第一个音

A velar sound: the first sound in cool

1.3 发音方式：塞音、擦音和近似音

1.3 Manner of Articulation: Stops, Fricatives and Approximants

我们现在已经确定了八个发音部位：声门、双唇、唇-牙、牙、牙槽、腭-牙槽、腭和软腭。对于任何给定的声音，我们会说它是浊音还是清音，以及它的发音位置。但是要区分语音的全部范围，我们将需要第三个描述性参数：发音方式。识别声音的方式发音时，我们将识别三种不同程度的收缩（完全闭合、接近闭合和闭合接近），从而识别出三种不同类别的辅音：塞音、摩擦音和近似音。

We have now identified eight places of articulation: glottal, bilabial, labio-dental, dental, alveolar, palato-alveolar, palatal and velar. For any given sound we will say whether it is voiced or voiceless, and what its place of articulation is. But to distinguish between the full range of speech sounds, we will require a third descriptive parameter: manner of articulation. To identify the manner in which a sound is articulated, we will identify three different degrees of constriction (complete closure, close approximation and open approximation), and thus three different categories of consonant: stops, fricatives and approximants.

1.3.1 停止

1.3.1 Stops

有问题的咬合架可能形成完全闭合的狭窄；这是当一个人在pit中发出第一个声音时发生的情况。在这里，下唇和上唇完全阻断了肺部的气流；然后可以释放关闭，因为它在坑中，然后可能会突然流出空气。完全闭合时发出的声音称为塞音（或爆破音）。

The articulators in question may form a stricture of complete closure; this is what happens when one produces the first sound in pit. Here the lower and upper lips completely block the flow of air from the lungs; that closure may then be released, as it is in pit, and may then produce a sudden outflow of air. Sounds which are produced with complete closure are referred to as stops (or plosives).

我们可以将pit中的第一个声音描述为清音双唇塞音（转录为 [p]），并且我们今后将识别所有具有此类三项标签的辅音。abbey中的辅音也是双唇塞音，但与pit中的不同：它是浊音。这个辅音（转录为 [b]）是浊双唇塞音。

We may describe the first sound in pit as a voiceless bilabial stop (transcribed as [p]) and we will henceforth identify all consonants with three-term labels of this sort. The consonant in abbey is also a bilabial stop, but differs from that in pit: it is voiced. This consonant (transcribed as [b]) is a voiced bilabial stop.

锡音中的第一个声音是清音牙槽塞音；它被转录为[t]。它的浊音对应物是ado中的辅音。这个声音，浊齿槽塞音，被转录为 [d]。

The first sound in tin is a voiceless alveolar stop; it is transcribed as [t]. Its voiced counterpart is the consonant in ado. This sound, the voiced alveolar stop, is transcribed as [d].

冷音中的第一个声音是清音软腭塞音；它被转录为[k]。它的浊音对应物，浊音软腭塞音，被转录为 [g]；一个例子是ago中的辅音。

The first sound in cool is a voiceless velar stop; it is transcribed as [k]. Its voiced counterpart, the voiced velar stop, is transcribed as [g]; an example is the consonant in ago.

我们现在已经确定了双唇、肺泡和软腭塞；停顿可以在许多其他发音处进行，但我们将忽略这些，因为它们与英语学习无关。然而，我们还必须提到一个停顿，因为它在大多数说英语的人的演讲中很常见。这是声门塞音（转录为 [ʔ]）。它是通过在声带之间形成完全闭合的收缩而制成的。这是许多苏格兰和伦敦发音中 [t] 的发音，例如单词butter的发音。我们将看到它出现在几乎所有说英语的人的讲话中，无论其口音如何。不存在将声门塞音描述为浊音或清音的问题，因为它是在声门本身中发音的。

We have now identified bilabial, alveolar and velar stops; stops may be made at many other places of articulation, but we will ignore those, as they are not relevant to the study of English. There is one further stop which we must mention, however, as it is very common in the speech of most speakers of English. This is the glottal stop (transcribed as [ʔ]). It is made by forming a constriction of complete closure between the vocal folds. This is the sound made instead of [t] in many Scottish and Cockney pronunciations of, for example, the word butter. We will see that it is present in the speech of almost every speaker of English, no matter what the accent. There is no question of describing the glottal stop as voiced or voiceless, since it is articulated in the glottis itself.

1.3.2 擦音

1.3.2 Fricatives

现在让我们区分完全闭合和另一种不太极端的收缩程度：近似。这种收缩产生的声音需要将两个咬合架结合到气流未完全阻塞的位置：留有足够的间隙让空气逸出，但咬合架靠得太近以至于产生摩擦随着空气逸出。这种声音被称为擦音。

Let us now distinguish between complete closure and another, less extreme, degree of constriction: close approximation. Sounds which are produced with this kind of constriction entail a bringing together of the two articulators to the point where the airflow is not quite fully blocked: enough of a gap remains for air to escape, but the articulators are so close together that friction is created as the air escapes. Sounds of this sort are referred to as fricatives.

fin中的第一个声音是通过将下唇靠近上牙以紧密近似的收缩来产生的。这种声音是无声的唇齿音擦音（转录为 [f]）。它的浊音对应物（浊唇齿擦音，记作 [v]）是Eva中的辅音。

The first sound in fin is created by bringing the lower lip close to the upper teeth in a constriction of close approximation. This sound is a voiceless labio-dental fricative (transcribed as [f]). Its voiced counterpart (the voiced labio-dental fricative, transcribed as [v]) is the consonant in Eva.

薄的第一个声音是通过使舌尖与上齿紧密接近而产生的。该音为清齿擦音，记为[θ]。它的浊音对应物，浊齿擦音（转录为 [ð]），对于某些说话者来说，是单词中的第一个声音。^2个

The first sound in thin is created by bringing the tip of the tongue into a constriction of close approximation with the upper teeth. This sound is a voiceless dental fricative, transcribed as [θ]. Its voiced counterpart, the voiced dental fricative (transcribed as [ð]) is, for some speakers, the first sound in the word that.2

sin中的第一个声音是通过将舌尖或舌叶带入与牙槽嵴紧密接近的收缩而产生的。此音记为 [s]，是清音牙槽擦音。它的浊音对应物，浊齿槽擦音（记为 [z]）是zoo中的辅音。

The first sound in sin is created by bringing the tip or blade of the tongue into a constriction of close approximation with the alveolar ridge. This sound, transcribed as [s], is a voiceless alveolar fricative. Its voiced counterpart, the voiced alveolar fricative (transcribed as [z]) is the consonant in zoo.

船上的第一个声音是通过将舌叶带入与腭-牙槽区域非常接近的收缩而产生的。此音记为 [ʃ]，是清腭-牙槽擦音。它的浊音对应物，转录为 [ʒ]，是seizure中的第二个辅音。

The first sound in ship is created by bringing the blade of the tongue into a constriction of close approximation with the palato-alveolar region. This sound, transcribed as [ʃ], is a voiceless palato-alveolar fricative. Its voiced counterpart, transcribed as [ʒ], is the second consonant in seizure.

摩擦音可以在任何发音点发出，但其中许多发音与英语学习无关。但是，我们会提到三个。

Fricatives may be articulated at any point of articulation, but many of those sounds are irrelevant to the study of English. However, we will mention three.

一种是清软颚擦音 [x]，出现在许多苏格兰人的语言中，如loch。另一个是清擦音 [ʍ]，在许多苏格兰人的语言中再次出现，例如whale（与wail相对）和which（与witch相对）；它的发音位置是唇软骨（在 1.3.3 中解释）。

One is the voiceless velar fricative [x], found in the speech of many Scots, in words such as loch. Another is the voiceless fricative [ʍ], again found in the speech of many Scots, as in words like whale (as opposed to wail) and which (as opposed to witch); its place of articulation is labial-velar (explained in 1.3.3).

第三个是声门擦音 [h]，如hit中的第一个音。这种声音是通过使声带紧密收缩而产生的，从而产生摩擦。由于声带不振动，我们会认为这是清音。

A third is the glottal fricative [h], as in the first sound in hit. This sound is produced by bringing the vocal cords into a constriction of close approximation, so that friction is produced. As the vocal cords are not vibrating, we will take it that this is a voiceless sound.

1.3.3 近似值

1.3.3 Approximants

当咬合架靠得相当近，但靠得不够近以产生摩擦时，会发生最轻微的收缩。这种狭窄称为开近似。以这种方式产生的辅音称为近似音。

The least radical degree of constriction occurs when the articulators come fairly close together, but not sufficiently close together to create friction. This kind of stricture is called open approximation. Consonants produced in this way are called approximants.

yes中的第一个音是近似音。它是通过将舌头前部靠近硬腭产生的。尽管舌头的两侧与上牙龈处于完全闭合的收缩状态，但空气会沿着中央凹槽逸出，舌头的前部在该凹槽中与硬腭的距离不足以产生摩擦。此音记为 [j]，是浊腭音。近似值通常是浊音，因此我们不会讨论这些声音的任何清音对应物。

The first sound in yes is an approximant. It is produced by bringing the front of the tongue close to the hard palate. Although the sides of the tongue are in a constriction of complete closure with the upper gums, the air escapes along a central groove in which the front of the tongue is not close enough to the hard palate to create friction. This sound, transcribed as [j], is a voiced palatal approximant. Approximants are normally voiced, so we will not discuss any voiceless counterparts for these sounds.

许多说英语的人对rip、rope、rat等的发音中的第一个音是近似音。它是通过使舌叶进入与牙槽嵴开放近似的收缩而产生的。这个近似值，转录为 [ɹ]，被称为肺泡近似值。与 [j] 一样，舌头的两侧与口腔两侧的牙龈形成完全闭合的收缩，但空气沿着中央凹槽逸出而不会产生摩擦。对于大多数说话者（并且在不同程度上取决于口音），在发出 [ɹ] 时舌体会稍微缩回；因此，它通常被称为肺泡后近似，但“肺泡近似”足以满足我们的目的。^3个

The first sound in many English speakers’ pronunciation of rip, rope, rat, etc. is an approximant. It is produced by bringing the blade of the tongue into a constriction of open approximation with the alveolar ridge. This approximant, transcribed as [ɹ], is referred to as an alveolar approximant. As with [j], the sides of the tongue form a constriction of complete closure with the gums at the sides of the mouth, but the air escapes along a central groove without creating friction. For most speakers (and in varying degrees, depending on the accent), the tongue body is somewhat retracted when [ɹ] is uttered; it is therefore often referred to as a post-alveolar approximant, but ‘alveolar approximant’ will suffice for our purposes.3

我们将在第 2 章中研究更多的英语近似音。现在，让我们再识别一个这样的声音，即wet开头的声音。在发出这种声音时，嘴唇形成了一种开放式的收缩：没有摩擦产生。但它的发音比 [j] 的发音更复杂，即上腭音，因为它还涉及舌背和软腭之间的另一个发音（即软腭发音）。因此，我们将其称为浊唇-软腭近似音；它被转录为[w]。

We will be looking at more English approximants in chapter 2. For the moment, let us identify one further such sound, the sound at the beginning of wet. In producing this sound, the lips form a constriction of open approximation: there is no friction produced. But its articulation is more complicated than that of [j], the palatal approximant, since it also involves another articulation, between the back of the tongue and the velum (i.e. a velar articulation). We will therefore refer to it as a voiced labial-velar approximant; it is transcribed as [w].

笔记

Notes

1许多音韵学家和语音学家使用术语“palato-alveolar”，但国际语音协会使用的符号表使用术语“post-alveolar”。如果学生认为这两个术语可以互换，那么对于我们的目的就足够了。牙槽嵴和硬腭之间没有严格的生理划分；从一个到另一个的过渡是一个连续体。两者之间的发音范围相对不同，导致一些语音学家将 alveo-patalal 和 palato-alveolar 发音区分开来。我们将通过忽略这些细节来简化。

1 Many phonologists and phoneticians use the term ‘palato-alveolar’, but the chart of symbols used by the International Phonetics Association uses the term ‘post-alveolar’. It will suffice for our purposes if the student takes the two terms to be interchangeable. There are no rigid physiological divisions between the alveolar ridge and the hard palate; the transition from one to the other is a continuum. And the range of articulations which can be made in between the two is relatively varied, leading some phoneticians to distinguish alveo-palatal from palato-alveolar articulations. We will simplify by ignoring these details.

2许多说英语的人没有浊齿擦音；相反，声音缺乏摩擦：它是一种浊音牙科近似音。

2 Many speakers of English do not have a voiced dental fricative; rather, the sound lacks friction: it is a voiced dental approximant.

3某些美国和西部乡村口音中 [ɹ] 类发音的发音也被一些人称为近似卷舌音。术语“舌尖”是指舌叶和舌尖向上和向后卷曲一定程度，从而使舌头的一部分的下侧形成相关的发音。有些不准确，我们将 [ɹ] 用于这些声音。

3 The articulation of an [ɹ] kind of articulation in some American and West Country accents is also referred to by some as retroflex approximant. The term ‘retroflex’ means that the blade and tip of the tongue are curled upwards and backwards to some extent, so that the underside of a part of the tongue forms the relevant articulation. Somewhat inaccurately, we will use [ɹ] for these sounds.

练习

Exercises

1 为以下每个声音给出适当的三项描述（例如 [k]：清软腭塞音）：

[θ] [b] [f] [ʃ] [j] [t]

2 为下列每个音标出相应的音标：

(a) 浊腭-牙槽擦音

(b) 浊齿槽塞音

(d) 浊齿擦音

(e) 浊唇齿擦音

3 以下每对声音的语音属性是什么（例如 [p] 和 [b]：发声；[s] 和 [ʃ]：发音位置；[t] 和 [s]：发音方式）？

(a) [k] 和 [g]

(b) [b]和[d]

(d) [z] 和 [ʒ]

(e) [ʃ] 和 [ʒ]

(f) [d] 和 [g]

gueqrecard_001

4在www.wiley.com/go/carrphonetics收听曲目 1.1。录音中的哪个单词以摩擦音开头？下面列出了这些词。

船舶心理学转向圆形情节哲学思考晚期木琴

5 听曲目 1.2。录音中哪个词以擦音结尾？下面列出了这些词。

栈鲸游泳墓志铭一半一半散列阴霾阶段使用路径削减恳求

6 听曲目 1.3。录音中的哪个词以停顿开头？下面列出了这些词。

花花公子泥水匠节俭的松鸡灵魂魅力谷物吹毛求疵古怪

的食尸鬼

7 描述发音时发音器官的位置和动作（例如 [d]：舌叶与牙槽嵴形成完全闭合的收缩；声带振动）：

[v] [θ] [k] [b]

2个

2 英语语音：辅音（二）

English Phonetics: Consonants (ii)

2.1 中央与横向

2.1 Central vs Lateral

在讨论肺泡近似 [ɹ] 时，我们说空气沿着中央凹槽逸出（在这种情况下是舌头；嘴唇可以形成相同类型的凹槽）。第 1 章中描述的所有擦音和逼近音都是如此：它们都是中心擦音和逼近音。然而，有可能产生并非如此的擦音和近似音。例如，在第一个升音中，舌叶的中心与部分牙槽嵴形成完全闭合的狭窄，但“重要”的发音是两侧之间的发音舌头和牙槽嵴。由于舌的两侧与牙槽嵴形成开放式发音的收缩，并且没有产生摩擦，我们将此声音（转录为 [l]）称为浊齿槽侧音。由于英语擦音和近似音通常是中央擦音，我们将使用术语“lateral”来表示侧音，并在描述英语语音中的中央擦音和近似音时省略术语“central”。发音 [l] 和 [ɹ] 显然非常相似：都是近似音，都是浊音，都是肺泡音。主要区别在于前者在侧面，而后者在中央。^1个

In discussing the alveolar approximant [ɹ], we said that the air escapes along a central groove (of the tongue, in this case; the same kind of groove can be formed by the lips). This is true for all of the fricatives and approximants described in chapter 1: they are all central fricatives and approximants. However, it is possible to produce fricatives and approximants in which this is not the case. For instance, in the first sound in lift, the centre of the blade of the tongue forms a stricture of complete closure with part of the alveolar ridge, but the articulation which ‘counts’ is that between the sides of the tongue and the alveolar ridge. Since the sides of the tongue form a constriction of open articulation with the alveolar ridge, and no friction is created, we refer to this sound (transcribed as [l]) as a voiced alveolar lateral approximant. Since English fricatives and approximants are typically central, we will use the term ‘lateral’ for laterals, and omit the term ‘central’ in describing central fricatives and approximants in English speech. The sounds [l] and [ɹ] are, clearly, quite similar: both are approximants, both are voiced, both are alveolar. The principal difference is that the former is lateral and the latter central.1

2.2 拍子和颤音

2.2 Taps and Trills

我们已经说过，对于很多说英语的人来说，rat、rope、reap等单词开头的声音是后牙槽音近似值：[ɹ]。诸如prude、true、creep等词中塞音后出现的声音也是如此。但是，有些说话者发出的声音不是近似音，而是非常像持续时间很短的浊音牙槽塞音。许多苏格兰人发出这种声音，而不是像刚刚引用的单词中的 [ɹ]，停顿之后。在发这个音的过程中，舌根暂时收缩，与牙槽嵴完全闭合。这种声音，转录为 [ɾ]，被称为浊音肺泡拍打声（或瓣膜）。这也是许多美国说话者在Betty、witty、rider、heady等词中代替 [t] 或 [d] 的声音。

We have said that, for a great many speakers of English, the sound at the beginning of words such as rat, rope, reap, etc. is a post-alveolar approximant: [ɹ]. The same is true of the sound which occurs after stops in words such as prude, true, creep, etc. However, some speakers utter, not an approximant, but a sound which is very like a voiced alveolar stop of very short duration. Many Scots utter this sound, rather than [ɹ], after stops, as in the words just cited. During the articulation of this sound, the blade of the tongue comes into a momentary constriction of complete closure with the alveolar ridge. This sound, transcribed as [ɾ], is referred to as a voiced alveolar tap (or flap). This is also the sound that many American speakers have instead of [t] or [d] in words such as Betty, witty, rider, heady, etc.

讲某些英语口音的人在rat、rope、reap和prude、true、creep等词中可能既不会发 [ɾ] 也不会发 [ɹ] ，但会发出一种被称为浊齿槽颤音的声音. 颤音是通过将一个咬合架（例如舌叶）紧挨着另一个咬合架（例如牙槽嵴）以完全闭合的收缩来产生的，但没有与停止时发现的相同的肌肉压力。结果是气压在封闭物后面积聚并迫使它打开；然后气压降低，肌肉压力再次产生完全闭合的收缩。这个序列可以快速连续地重复，在牙槽颤音的情况下，产生一系列舌头敲击牙槽嵴的声音。肺泡颤音被转录为 [r]，但相对少见。人们常说苏格兰人会发出这种声音；然而，大多数讲苏格兰英语的人通常发出的不是肺泡颤音，而是肺泡拍打声。

Speakers of certain accents of English may utter neither an [ɾ] nor an [ɹ] in words such as rat, rope, reap and prude, true, creep, but a sound referred to as a voiced alveolar trill. Trills are produced by holding one articulator (e.g. the blade of the tongue) next to the other (e.g. the alveolar ridge) in a constriction of complete closure, but without the same muscular pressure as one finds in stops. The result is that air pressure builds up behind the closure and forces it open; the air pressure then reduces, and the muscular pressure again creates a constriction of complete closure. This sequence may be repeated in quick succession, producing, in the case of an alveolar trill, a series of taps of the tongue against the alveolar ridge. The alveolar trill is transcribed as [r], but is relatively rare. Scots are often said to produce this sound; however, most speakers of Scottish varieties of English typically produce, not an alveolar trill, but an alveolar tap.

2.3 二次发音

2.3 Secondary Articulation

我们已经说过，横向近似值 [l] 是肺泡的。然而，侧音也可以通过额外的发音产生，例如在舌背和软腭之间形成的发音，即软腭发音。当发生这种情况时，我们可以将牙槽骨关节区分为主要关节和软骨关节作为次要关节。如果第二个关节是软骨，这个过程被称为软骨化：我们说侧面是软骨化的。软腭化的横向近似词使用软腭化变音符号转录，因此：[ɫ]。这种声音通常被称为“dark l”。^2个如果次要发音是腭音（在舌头前部和硬腭之间形成），则此过程称为腭化；我们说侧音是腭化的。腭化侧音使用腭化变音符号转录，因此：[l ^j ]。术语“clear l”通常用于指代 [l ^j ] 或 [l]（既不腭化也不“暗”）。在随后的章节中，我们将考虑“dark l”和“clear l”在不同英语口音中的地位。

We have said that the lateral approximant [l] is alveolar. However, laterals may also be produced with an additional articulation, such as one formed between the back of the tongue and the velum, i.e. a velar articulation. When this happens, we may distinguish between the alveolar articulation as the primary articulation and the velar one as the secondary articulation. Where a secondary articulation is velar, this process is referred to as velarization: we say that the lateral is velarized. A velarized lateral approximant is transcribed using the velarization diacritic, thus: [ɫ]. This sound is often referred to as ‘dark l’.2 Where a secondary articulation is palatal (formed between the front of the tongue and the hard palate), this process is referred to as palatalization; we say that the lateral is palatalized. A palatalized lateral is transcribed using the palatalization diacritic, thus: [lj]. The term ‘clear l’ is often used to refer to [lj], or to [l] (neither palatalized nor ‘dark’). In subsequent chapters, we will consider the status of ‘dark l’ and ‘clear l’ in different accents of English.

2.4 附属词

2.4 Affricates

到目前为止，我们已经根据收缩程度区分了三类辅音：塞音、擦音和近似音。考虑chip中的第一个声音：它就像一个停止，因为舌叶和腭牙槽区域之间完全闭合。然而，它就像一个摩擦音，因为它显然包含摩擦。这种摩擦发生在我们在 1.3.1 中提到的闭包的释放阶段。完全闭合收缩时产生的声音随后是发生摩擦的释放阶段，称为塞擦音。我们可能会说，tip中的第一个声音和chip中的第一个声音之间的主要区别之一（发音位置分开）是，在tip中 [t] 的释放阶段，没有那种摩擦在chip的第一个声音的释放阶段发现。因此，我们可能会将塞擦音视为具有缓慢、摩擦音的释放阶段的塞音。芯片中的塞擦音是清音腭牙槽塞擦音，转录为 [tʃ]。它的浊音对应物是 [dʒ]，jury、joy等中的第一个音³

We have, thus far, distinguished three classes of consonant according to degree of constriction: stops, fricatives and approximants. Consider the first sound in chip: it is like a stop in that there is complete closure between the blade of the tongue and the palato-alveolar region. However, it is like a fricative in that it clearly involves friction. That friction occurs during the release phase of the closure, which we referred to in 1.3.1. Sounds produced with a constriction of complete closure followed by a release phase in which friction occurs are called affricates. We might say that one of the main differences (place of articulation apart) between the first sound in tip and the first sound in chip is that, during the release phase of the [t] in tip, there is no friction of the sort one finds during the release phase of the first sound in chip. We might therefore think of affricates as stops with a slow, fricative, release phase. The affricate in chip is a voiceless palato-alveolar affricate, transcribed as [tʃ]. Its voiced counterpart is [dʒ], the first sound in jury, joy, etc.3

这两个塞擦音出现在大多数说英语的人的演讲中。在后面的章节中，我们将研究出现在说某些英语口音的人说话时出现的其他一些塞擦音。

These two affricates occur in the speech of most speakers of English. In later chapters, we will examine some other affricates which occur in the speech of speakers of certain accents of English.

2.5 抱负

2.5 Aspiration

我们说过， pit的第一站是清音双唇站。所以也是第一站吐。但是，pit中的双唇塞音与 spit中的双唇塞音在语音上有所不同：如果你在吐pit时将手掌靠近嘴巴，那么在释放双唇塞音时，你会感觉到比你吐出时更强烈的气流。彻底吐了这种“更强的气流”现象称为送气：我们说双唇凹陷中的塞音是吸气清音塞，而吐出的塞音是不送气的. 送气清塞音用送气变音符号 ([ʰ]) 转录，因此双唇塞音pit被转录为 [pʰ]。不送气的塞音在没有变音符号的情况下转录，因此spit中的双唇塞音被转录为 [p]。

The first stop in pit, we said, is a voiceless bilabial stop. So too is the first stop in spit. But the bilabial stop in pit differs phonetically from the bilabial stop in spit: if you hold the palm of your hand up close to your mouth when uttering pit, you will feel a stronger puff of air on releasing the bilabial stop than you will when you utter spit. That ‘stronger puff of air’ phenomenon is called aspiration: we say that the bilabial stop in pit is an aspirated voiceless stop, whereas the stop in spit is unaspirated. Aspirated voiceless stops are transcribed with the aspiration diacritic ([ʰ]), so that the bilabial stop in pit is transcribed as [pʰ]. Unaspirated stops are transcribed without that diacritic, so that the bilabial stop in spit is transcribed as [p].

2.6 鼻塞

2.6 Nasal Stops

到目前为止，我们一直在我们的讨论中做出假设，关于软膜在我们所描述的语音产生中的位置。我们假设，在所有这些声音中，肺部的空气仅通过口腔（口腔）逸出。如果软帆处于升高位置，则情况确实如此，这样它可以防止空气从鼻腔流出（见图 1）。在迄今为止讨论的所有声音中，确实提高了音调：我们将所有这些声音描述为口音。但可以降低帆板，以允许空气通过鼻腔逸出（见图 1）。软帆降低而空气单独通过鼻腔逸出时产生的声音被称为鼻塞。⁴这些可能出现在大多数发音部位；让我们考虑那些与英语学习相关的问题。

We have been making an assumption in our discussion thus far, concerning the position of the velum in the production of the speech sounds we have described. We have assumed that, in all of these sounds, the air from the lungs is escaping only through the mouth (the oral cavity). This is true if the velum is in the raised position, such that it prevents the flow of air out through the nasal cavity (see figure 1). In all of the sounds discussed thus far, the velum is indeed raised: we describe all such sounds as oral sounds. But the velum may be lowered, to allow escape of air through the nasal cavity (see figure 1). Sounds produced with the velum lowered, and with air escaping through the nasal cavity alone, are referred to as nasal stops.4 These may occur at most places of articulation; let us consider those which are relevant for the study of English.

虽然鼻塞音可能是浊音或清音，但在大多数人类语言中它们通常是浊音；因此，我们将忽略清鼻音并使用术语“鼻音”来暗示“浊鼻音”。

While nasal stops may be either voiced or voiceless, they are typically voiced in most human languages; we will therefore ignore voiceless nasal stops and use the term ‘nasal stop’ to imply ‘voiced nasal stop’.

双唇鼻塞（转录为 [m]）正如人们所预料的那样，需要双唇完全闭合、发声和空气通过鼻腔逸出。一个例子是map中的第一个辅音。

Bilabial nasal stops (transcribed [m]) entail, as one would expect, complete closure between the lips, voicing, and escape of the air through the nasal cavity. An example is the first consonant in map.

Labio-dental 鼻塞（转录为 [ɱ]）需要下唇和上牙之间的完全闭合、发声和空气通过鼻腔逸出。一个例子是小册子中的第二个辅音。在英语中，它们出现在 labio-dental 音之前，就像在本例中一样。这些情况下的鼻塞发音反映了同化过程。同化过程是一个声音变得与相邻声音相似的过程。在这种情况下，鼻音被同化为后面的擦音，因为它“取代”了擦音的发音。这种过程涉及易于表达的原则。在这种情况下，如果小册子中的鼻音与后面的擦音发音相同的地方，这样可以节省说话者从双唇发音到唇齿发音的发音努力。我们将在第 6 章回到这些过程。

Labio-dental nasal stops (transcribed [ɱ]) entail complete closure between the lower lip and the upper teeth, voicing, and escape of the air through the nasal cavity. An example is the second consonant in pamphlet. In English, they occur before labio-dental sounds, as in this case. The nasal stop articulation in cases such as these reflects a process of assimilation. Assimilation processes are processes in which one sound becomes similar to an adjacent sound. In this case, the nasal is assimilated to the following fricative, in the sense that it ‘takes on’ the place of articulation of the fricative. Such processes involve a principle of ease of articulation. In this case, if the nasal in pamphlet is articulated at the same place as the following fricative, this saves the speaker the articulatory effort of moving from a bilabial to a labio-dental articulation. We will return to such processes in chapter 6.

牙科鼻塞（转录为 [n̪]）需要舌尖和上牙之间的完全闭合、发声和空气通过鼻腔逸出。一个例子是第十个中的第二个辅音。在这种情况下，它们出现在其他牙音之前，这也是涉及发音位置的同化问题。

Dental nasal stops (transcribed as [n̪]) entail complete closure between the tip of the tongue and the upper teeth, voicing, and escape of the air through the nasal cavity. An example is the second consonant in tenth. As in this case, they occur before other dental sounds, and this too is a matter of assimilation involving place of articulation.

牙槽鼻塞（转录为 [n]）需要舌叶和牙槽脊之间的完全闭合、发声和空气通过鼻腔逸出。一个例子是not中的第一个声音。

Alveolar nasal stops (transcribed as [n]) entail complete closure between the blade of the tongue and the alveolar ridge, voicing, and escape of the air through the nasal cavity. An example is the first sound in not.

软腭鼻塞（转录为 [ŋ]）需要舌后部和软帆之间的完全闭合、发声和空气通过鼻腔逸出。一个例子是sing中的最后一个音或鼻塞音，因为它经常在单词incredible 中清晰表达（尤其是在更快或更随意的演讲风格中）。后一种情况再次涉及同化。

Velar nasal stops (transcribed as [ŋ]) entail complete closure between the back of the tongue and the velum, voicing, and escape of the air through the nasal cavity. An example is the last sound in sing or the nasal stop as it is often articulated (especially in faster or more casual speech styles) in the word incredible. Once again, the latter case involves assimilation.

笔记

Notes

1中央逼近音 [ɹ] 与 [l] 的区别还在于舌体后缩和唇部变圆。我们很快就会看到，牙槽支管也可能在收缩时产生。

1 The central approximant [ɹ] also differs from [l] in having tongue body retraction and lip rounding. We will see shortly that alveolar laterals may be produced with retraction too.

2术语“暗 l”也可用于指代舌体/后部缩回和/或降低的侧向近似。英语口音在其“dark l”的确切发音性质方面有所不同：有些是软腭化的，而另一些则没有软腭发音，而是有缩回和/或降低舌背/舌体。这种回缩会导致肺泡接触的丧失，从而导致 [l]-发声，其中发音变得像元音一样。

2 The term ‘dark l’ can also be used to refer to lateral approximants in which the body/back of the tongue is retracted and/or lowered. Accents of English vary with respect to the exact articulatory nature of their ‘dark l’s: some are velarized, while others have no velar articulation, but have, instead, retraction and/or lowering of the back/body of the tongue. Such retraction can lead to loss of alveolar contact, and thus to [l]-vocalization, in which the articulation becomes vowel-like.

3一些作者将 [tʃ] 转录为 [č]，将 [dʒ] 转录为 [ǰ]。如果我们要严格遵守国际语音协会的惯例，我们应该在两个符号上方用“连杆”来转录两个塞克音；我们在这里偏离了 IPA 图表的惯例，该图表不包含“擦除”类别。

3 Some authors transcribe [tʃ] as [č] and [dʒ] as [ǰ]. We should, if we were to stick strictly to the conventions of the International Phonetics Association, transcribe both affricates with a ‘tie bar’ above the two symbols; we depart here from the conventions of the IPA chart, which does not contain an ‘affricate’ category.

4术语“鼻化”与“鼻音”相对，用于描述空气通过口腔和鼻腔逸出的声音。术语“鼻音”用于描述空气仅通过鼻腔逸出的声音。

4 The term ‘nasalized’, as opposed to ‘nasal’, is used to describe sounds in which air escapes through both cavities, the oral and the nasal. The term ‘nasal’ is used to describe sounds in which the air escapes through the nasal cavity alone.

练习

Exercises

gueqrecard_001

1在www.wiley.com/go/carrphonetics收听曲目 2.1。对于录音中的每个单词，确定 (a) 任何口头塞音，(b) 任何擦音，(c) 任何近似音，(d) 任何塞擦音和 (e) 任何鼻音。对于您识别的每个声音，说出它是浊音还是清音，以及它的发音位置（例如，单词stop：清音齿龈塞音 [t] 和清双唇音 [p]；清音齿龈擦音 [s]；无近似词、塞擦音或鼻音）。话是：

带来舔热思维保证措施加热多年蠕虫丛林

2 听曲目 2.2。录音中的哪些单词以塞擦音开头，哪些（如果有的话）以塞音开头？

调整芯片沙丘六月

许多说英语的人通常会在单词开头加上塞擦音，比如tune和dune 。这意味着沙丘和六月通常无法区分。尽管如此，当在语音课上被问到他们是否会用塞擦音说出诸如沙丘之类的词时，这些说话者通常会否认他们这样做。这些说话者通常对诸如tune和dune之类的词有更仔细的发音，其中有一个[j]。

Many speakers of English typically utter words like tune and dune with an affricate at the beginning of the word. This means that dune and June are typically indistinguishable. None the less, when asked in a phonetics class whether they utter words such as dune with an affricate, such speakers often deny that they do. These speakers typically have a more careful pronunciation of words such as tune and dune, in which there is a [j].

但是请注意，没有像chip和June这样更仔细的单词发音：人们从来没有听过 [tj] 和 [dj] 的发音。为了解释dune和June之间的区别，我们需要说说话者在某种意义上打算在dune中发声 [dj] ，但这种发音的轻松导致塞音闭合的腭-牙槽塞音释放，而不是舌头前部和硬腭之间从肺泡闭合到开放近似狭窄的过渡。在June和chip的情况下，预期的发音是腭牙槽塞音。

Notice, however, that there is no such more careful pronunciation of words like chip and June: one never hears these pronounced with [tj] and [dj]. In order to explain the difference between dune and June, we need to say that the speaker in some sense intends to utter [dj] in dune, but that ease of articulation results in a palato-alveolar affricated release of the stop closure, rather than a transition from an alveolar closure to a stricture of open approximation between the front of the tongue and the hard palate. In the case of June and chip, the intended articulation is a palato-alveolar affricate.

如果您说通用美国语，您可能永远不会在tune和dune等词中发出 [j] ，在这种情况下，您会发出塞音后跟一个元音。但是，您可能在学校也被告知，此类单词的“正确”发音在塞音后有一个 [j]。关于 [j] 的存在与否，您的演讲可能会有所不同。如果您的讲话确实以这种方式变化，您如何发音noon？

If you are a speaker of General American, you may well never utter a [j] in words like tune and dune, in which case you will utter a stop followed by a vowel. However, you may well also have been told at school that the ‘correct’ pronunciation of such words has a [j] after the stop. Your speech may well vary with respect to the presence or absence of the [j]. If your speech does vary in this way, how do you pronounce noon?

gueqrecard_002

3 听曲目 2.3。给出录音中每个单词的音标，使用“V”表示元音。话是：

lull pear reap throws think misjudges churches 不正确的输入

你可能已经注意到think中的鼻塞是软腭的，而不是肺泡的。将鼻音塞音发音为牙槽音需要相当大的有意识的努力，而且当一个人这样做时，由此产生的发音听起来很不自然。这似乎是预期同化过程的结果：在发鼻音时，舌头采用软腭塞音 [k] 的发音位置。

You may well have noticed that the nasal stop in think is velar, rather than alveolar. It requires considerable conscious effort to utter that nasal stop as alveolar, and when one does so, the resulting pronunciation sounds quite unnatural. This appears to be the result of a process of anticipatory assimilation: the tongue adopts the articulatory position for the velar stop [k] during the pronunciation of the nasal.

但是鼻加软腭停止顺序不正确怎么办？许多说英语的人发现在那里更容易说出牙槽音而不是软骨音，尽管事实上像不正确的情况也包含一系列鼻塞音和软骨塞音。对于为什么这两种情况应该不同，您有什么预感吗？

But what about the nasal-plus-velar stop sequence in incorrect? Many speakers of English find it easier to utter an alveolar, rather than a velar, nasal there, despite the fact that cases like incorrect also contain a sequence of a nasal stop followed by a velar stop. Do you have any hunches as to why the two cases should be different?

3个

3 英语语音：元音 (i)

English Phonetics: Vowels (i)

3.1 基本元音

3.1 The Primary Cardinal Vowels

让我们首先假设所有元音都是浊音，并以开放式近似的收缩发音。目前，我们还将假设所有元音都是口音（即，在它们的产生过程中提高了软膜）。舌头在口腔内可以占据的位置范围相当大，同时保持在开放近似的收缩中。让我们将此类发音的整个可用空间称为元音空间。我们将需要一种方法来绘制给定元音在元音空间中的发音点。为了做到这一点，我们将求助于该空间的理想化图表，如下所示（该图表在图 2的 IPA 图表中重复出现）：

Let us begin by assuming that all vowels are voiced and are articulated with a constriction of open approximation. We will also assume, for the moment, that all vowels are oral sounds (i.e. that the velum is raised during their production). The range of positions which the tongue can occupy within the oral cavity while remaining in a constriction of open approximation is quite large. Let us call the entire available space for such articulations the vowel space. We will require a means of plotting the point at which a given vowel is articulated in the vowel space. In order to do this, we will appeal to an idealized chart of that space, as follows (this chart is repeated in the IPA chart in figure 2):

(1) 元音空间与原基元音

在此图中，我们表示沿二维的元音空间。第一个是高/低维度（也称为闭/开维度），描述元音发音时舌体的高度（即描述元音高度）。这在图中表示为垂直轴。第二个是前/后维度，描述了舌体朝向元音空间前方的程度。这在图中表示为水平轴。我们可以沿着这个维度确定三个任意点：前面、中央和后面. 在使用这两个维度时，我们可以说，对于任何给定的元音，它在元音空间中的发音有多高，以及它是前元音、中元音还是后元音。对于这两个描述性参数，我们将添加第三个，它指的是嘴唇位置：我们将说，对于给定的元音，在其发音过程中，嘴唇是否圆润。我们将前一种元音称为圆唇元音，将后者称为不圆唇元音。

In this diagram, we represent the vowel space along two dimensions. The first is the high/low dimension (also referred to as the close/open dimension), depicting the height of the body of the tongue during the articulation of a vowel (i.e. depicting vowel height). This is represented as the vertical axis in the diagram. The second is the front/back dimension, depicting the extent to which the body of the tongue lies towards the front of the vowel space. This is represented as the horizontal axis in the diagram. We may identify three arbitrary points along this dimension: front, central and back. In using these two dimensions, we can say, for any given vowel, how high in the vowel space it is articulated, and whether it is a front, central or back vowel. To these two descriptive parameters, we will add a third, which refers to lip position: we will say, for a given vowel, whether, during its articulation, the lips are rounded or not. We will refer to the former sort of vowel as a rounded vowel and the latter as an unrounded vowel.

沿着元音空间的周边识别几个点很方便。完成此操作后，我们可以绘制任何给定元音相对于这些点的位置。在这些点发出的元音被称为基数元音。我们现在将确定其中的八个。

It is convenient to identify several points along the perimeter of the vowel space. Once we have done this, we can plot the location of any given vowel in relation to those points. Vowels articulated at those points are called the cardinal vowels. We will now identify eight of them.

让我们从嘴唇不圆润时产生的元音开始，舌头在元音空间中尽可能高和尽可能靠前，不会引起摩擦。这是基数元音 no。如图1所示，在上面 (1) 中的图表的左上角描绘。该元音被转录为 [i]。使用我们的三个描述性参数，我们将其称为前高不圆唇元音。我们不会试图用英语或任何其他语言的单词来举例说明基元音，因为通常说话者不会发出与基元音一样在元音空间中非常外围的元音。相反，我们将绘制英语元音相对于基元音，使用元音空间图作为元音空间的映射。例如，许多说英语的人对单词peep的发音中的元音与基数元音 no 非常接近。1：它也是一个高前圆唇元音，但不像基数元音 no.那样外围。1：它的发音通常略低一些，略微靠前一些。

Let us begin with the vowel which is produced when the lips are unrounded and the tongue is located as high as possible and as front as possible, without causing friction, in the vowel space. This is cardinal vowel no. 1, depicted at the top left-hand corner of the diagram in (1) above. That vowel is transcribed as [i]. Using our three descriptive parameters, we will refer to this as a high front unrounded vowel. We will not seek to exemplify cardinal vowels with words from English, or any other language, since, typically, speakers do not utter vowel sounds which are quite as peripheral in the vowel space as the cardinal vowels. Rather, we will plot the place of articulation of English vowels in relation to the cardinal vowels, using the vowel space diagram as a map of the vowel space. The vowel in many English speakers’ pronunciation of the word peep, for instance, is quite close to cardinal vowel no. 1: it too is a high front rounded vowel, but it is not quite as peripheral as cardinal vowel no. 1: it is typically slightly less high and slightly less front in its articulation.

现在让我们确定位于元音空间“另一端”的基数元音：当嘴唇不圆且舌体尽可能低和尽可能靠后时产生的元音，不会引起摩擦。这是基数元音 no。5 . 它的位置描绘在上面 (1) 中图表的右下角。音译为[ɑ]，为低后不圆唇元音。

Let us now identify the cardinal vowel which lies at the ‘opposite end’ of the vowel space: the vowel which is produced when the lips are unrounded and the body of the tongue is as low as possible and as far back as possible, without causing friction. This is cardinal vowel no. 5. Its location is depicted at the bottom right-hand corner of the diagram in (1) above. Transcribed as [ɑ], it is a low back unrounded vowel.

我们现在已经确定了元音空间中的两个“锚点”；我们现在可以继续确定与这些相关的更多基数元音。如果嘴唇保持不圆，舌体在元音空间中保持尽可能低（对于第 5 个基数元音），但舌头在不引起摩擦的情况下尽可能远地移动到该空间的前面，然后基数元音没有。4产生。它被转录为[a]。

We have now identified two ‘anchor’ points in the vowel space; we may now proceed to identify further cardinal vowels in relation to these. If the lips remain unrounded and the body of the tongue remains as low as possible in the vowel space (as for cardinal vowel no. 5), but the tongue is moved as far to the front of that space as is possible without causing friction, then cardinal vowel no. 4 is produced. It is transcribed as [a].

我们现在已经确定了两个元音高度：高和低。如果你说基数元音 no，你应该能够感觉到舌头高度的这种差异。1 后接基数元音 no. 4：当一个人从前者移动到后者时，下颌明显张开，舌体明显降低。在这两个高度之间有一个连续的元音高度；我们将沿着这个连续统一体确定两个任意点：中高点和中低点。如果嘴唇保持不圆，舌体保持尽可能靠前，但舌头高度比基数元音 no. 略低。1 位置，一个到达前面，高中不圆唇元音，称为基本元音 1。2 . 这被转录为[e]。

We have now identified two vowel heights: high and low. You should be able to feel this difference in tongue height if you utter cardinal vowel no. 1 followed by cardinal vowel no. 4: the jaw opens considerably and the body of the tongue lowers considerably as one moves from the former to the latter. There is a continuum of vowel heights between these two heights; we will identify two arbitrary points along this continuum: high-mid and low-mid. If the lips remain unrounded and the body of the tongue remains as far front as is possible, but the tongue height is lowered somewhat from the cardinal vowel no. 1 position, one arrives at the front, high-mid unrounded vowel known as cardinal vowel no. 2. This is transcribed as [e].

在保持相同的嘴唇位置和相同的前倾度的情况下，可以将舌体进一步降低到中低位置，并到达前低中不圆唇元音，称为基数元音 1。3 . 这被转录为 [ε]。

In retaining the same lip position and the same degree of frontness, one may lower the body of the tongue further still to the low-mid position, and arrive at the front low-mid unrounded vowel known as cardinal vowel no. 3. This is transcribed as [ε].

如果你说出基数元音没有。1，然后是基数元音 nos。2、3、4，你应该感觉到舌体逐渐下降。这些都是前不圆唇元音：它们之间的区别在于舌头的高度。

If you articulate cardinal vowel no. 1, then cardinal vowels nos. 2, 3 and 4, you should feel the body of the tongue lowering progressively. These are all front unrounded vowels: the difference between them lies in the height of the tongue.

现在让我们考虑后基元音。如果舌体在不引起摩擦的情况下尽可能高并尽可能向后，并且这次嘴唇是圆的，那么基数元音 no. 产生了8个。这个高后圆唇元音被转录为 [u]。

Let us now consider the back cardinal vowels. If the body of the tongue is as high as possible and as far back as possible without causing friction, and the lips are, this time, rounded, then cardinal vowel no. 8 is produced. This high back rounded vowel is transcribed as [u].

如果嘴唇保持圆润，舌头尽可能向后，但舌头高度降低到中高位置，基元音 no. 7产生。这个高中后圆唇元音被转录为 [o]。

If the lips remain rounded and the tongue remains as far back as possible, but the tongue height is lowered to the high-mid position, cardinal vowel no. 7 is produced. This high-mid back rounded vowel is transcribed as [o].

在保持相同的靠背度数和相同的嘴唇位置的情况下，人们可能会进一步降低舌头的高度，降低到中低位置，从而产生被称为基本元音1 的中低后圆唇元音。6 . 这被转录为 [ɔ]。

In retaining the same degree of backness and the same lip position, one may lower the height of the tongue still further, to the low-mid position, and thus produce the low-mid back rounded vowel known as cardinal vowel no. 6. This is transcribed as [ɔ].

当您从第 1 号基数元音过渡时，您应该能够感觉到舌头逐渐降低。8，通过基数元音没有。7，基数元音没有。6; 舌头经历与从第 1 号基数元音过渡时相同的降低过程。1，通过没有。2，没有。3.

You should be able to feel the tongue lowering progressively as you make the transition from cardinal vowel no. 8, through cardinal vowel no. 7, to cardinal vowel no. 6; the tongue goes through the same lowering process as it does for the transition from cardinal vowel no. 1, through no. 2, to no. 3.

我们现在已经确定了八个基本元音。建立这些参考点后，我们可以描述与它们相关的特定英语元音的发音。让我们先看看那些被称为英语短元音的字母。

We have now identified the eight primary cardinal vowels. With these reference points established, we may describe the articulation of specific English vowels in relation to them. Let us begin by looking at those referred to as the English short vowels.

3.2 RP 和 GA 短元音

3.2 RP and GA Short Vowels

不同英语口音的人发出的元音有相当大的差异，我们将在后面的章节中考虑这种差异。目前，我们将从两种特殊的口音开始；我们稍后会描述其他的。我们会有点武断地从被称为标准发音 (RP)和通用美国发音 (GA)的口音开始。RP 是经常提到的口音作为英国社会的“声望”口音，并与英国公立学校毕业生的演讲有关。因此，它主要根据说话者的社会阶层来定义。出于这个原因，我们不选择它作为我们的起点之一；相反，我们选择它是因为它往往是外国学习英国英语的人所学的口音，因此被广泛描述。GA 倾向于根据其说话者的地理位置而不是社会阶层来定义。“GA”一词是对居住在美国大部分地区的一组口音的理想化：它不包括纽约市口音等东方口音和德克萨斯州口音等南方口音。

There is considerable variation in the vowel sounds uttered by speakers of different accents of English, and we will be considering that variation in later chapters. For the moment, we will begin with two particular accents; we will later describe others. We will, somewhat arbitrarily, begin with the accents known as Received Pronunciation (RP) and General American (GA). RP is the accent often referred to as the ‘prestige’ accent in British society and associated with the speech of the graduates of the English public schools. It is thus defined largely in terms of the social class of its speakers. We do not select it as one of our starting points for that reason; rather, we select it as it tends to be the accent which foreign learners of British English are taught, and has thus been widely described. GA tends to be defined in terms of the geographical location, rather than the social class, of its speakers. The term ‘GA’ is an idealization over a group of accents whose speakers inhabit a vast proportion of the United States: it excludes Eastern accents such as the New York City accent, and Southern accents (such as that spoken in Texas).

人们经常指出，诸如“RP”和“GA”之类的术语需要大量的理想化，因为它们被用来涵盖各种不同但趋同的口音。我们认为这是不可避免的：我们用来描述口音的任何术语（例如“纽约市”、“Cockney”、“Scouse”、“Geordie”、“South African”等）都是如此，事实上“口音”一词本身也是如此。但是我们需要某种方式来表达关于不同语音社区成员发出的语音的有效概括。例如，通常情况下，虽然 RP 扬声器发音为put和putt不同的是，许多在英格兰北部发现有口音的人却没有。拒绝谈论不同的口音无异于把婴儿连同洗澡水一起倒掉，并且如果受到某些警告的约束，就剥夺了我们自己表达信息性陈述的机会。

It has often been pointed out that terms such as ‘RP’ and ‘GA’ entail a great deal of idealization, in that they are used to cover a variety of somewhat different, if converging, accents. We accept this as inevitable: it will be true of any term we use to describe an accent (e.g. ‘New York City’, ‘Cockney’, ‘Scouse’, ‘Geordie’, ‘South African’, etc.) and indeed it is true of the term ‘accent’ itself. But we need some way of expressing valid generalizations about the speech sounds which members of different speech communities utter. For instance, it is generally true that, while RP speakers pronounce put and putt differently, many speakers with accents found in the North of England do not. To refuse to speak of different accents would be to throw the baby out with the bathwater, and to deny ourselves the opportunity of expressing statements which are informative, if subject to certain caveats.

到目前为止，我们还没有谈及元音的长度。对于说 RP 和 GA 的人，peep和pip中的元音在几个方面有所不同，其中之一就是元音长度。如果您是 RP 或 GA 使用者，并且您说出这两个词，您可能会同意前者的元音比后者长。因此，我们将前者称为长元音，将后者称为短元音。元音长度是一个相对的问题：当我们说pip中的元音是短元音时，我们并不是指它的持续时间（以毫秒为单位）；相反，我们是说它比其他元音短，例如peep中的元音。pip中的元音通常发音时舌体相当靠前且相当高，嘴唇不圆。我们将该元音转录为 [ɪ]。虽然它是一个高前不圆唇元音，但它比 peep 中的元音低且前部少。其位置如下面 (2) 所示。

We have said nothing, as yet, about the length of vowels. For speakers of RP and GA, the vowels in peep and pip differ in several respects, one of which is vowel length. If you are an RP or a GA speaker, and you utter the two words, you will probably agree that the vowel in the former is longer than that in the latter. We will, accordingly, refer to the former as a long vowel and the latter as a short vowel. Vowel length is a relative matter: when we say that the vowel in pip is a short vowel, we are not referring to its duration in milliseconds; rather, we are saying that it is short in relation to other vowels, such as that in peep. The vowel in pip is typically articulated with the body of the tongue fairly front and fairly high, and with the lips unrounded. We will transcribe that vowel as [ɪ]. While it is a high front unrounded vowel, it is less high and less front than the vowel in peep. Its location is depicted in (2) below.

现在考虑 RP 和 GA 说话者对单词put中元音发音的元音。对于许多说话者来说，这是一个高后圆唇元音，在接近基本元音 1 的区域发音。8. 类似于学校的元音，但少高少回。也比学校短。我们将这个短元音转录为[ʊ]；它的位置在下面的 (2) 中描述。

Now consider the vowel in RP and GA speakers’ pronunciation of the vowel in the word put. This is, for many speakers, a high back rounded vowel, articulated in the region near to cardinal vowel no. 8. It is similar to the vowel in school, but less high and less back. It is also shorter than that in school. We will transcribe this short vowel as [ʊ]; its location is depicted in (2) below.

对于 RP 和 GA 说话者， put和putt中的元音是有区别的。两者都是短元音，但它们在几个方面有所不同。首先，后面的元音是不圆唇的。其次，putt中的元音发音时舌头高度相当低：通常，它刚好低于中低位置。第三，推杆中的元音位于前后轴的中点附近。我们将把位于该区域的元音称为中央元音。我们将这个元音转录为[ʌ]；它的位置在下面的 (2) 中描述。

For RP and GA speakers, there is a distinction between the vowel in put and that in putt. Both are short vowels, but they differ in several respects. Firstly, the latter vowel is unrounded. Secondly, the vowel in putt is articulated with a fairly low tongue height: typically, it is just below the low-mid position. Thirdly, the vowel in putt is located at around the half-way point on the front/back axis. We will refer to vowels located in this region as central vowels. We will transcribe this vowel as [ʌ]; its location is depicted in (2) below.

在 RP 和 GA 中，aunt和ant中的元音不同。两个元音均不圆唇，但ant的元音比aunt短，ant为低前元音，aunt为低后元音。ant中的前低不圆唇元音比基数元音 no. 发音得更高，更靠前。4. 我们将其转录为 [æ]；它的位置在下面的 (2) 中描述（尽管 GA 元音高于 RP 元音，并且对英国人来说听起来更像 [ε]）。

In both RP and GA, the vowels in aunt and ant differ. Both vowels are unrounded, but the vowel in ant is shorter than that in aunt, and the vowel in ant is a low front vowel, whereas that in aunt is a low back vowel. The low front unrounded vowel in ant is articulated higher and less front than cardinal vowel no. 4. We will transcribe this as [æ]; its location is depicted in (2) below (although the GA vowel is higher than the RP vowel, and sounds rather [ε]-like to British speakers).

RP和GA发音中的短元音是一个前不圆唇元音，其高度介于基数元音nos之间。2 和 3。对于大多数 RP 和 GA 说话者，它更接近基数元音 no。3比基数元音没有。2 高；它也比基数元音 no 更集中一些。3. 为方便起见，我们将其记为[ε]；它的位置在下面的 (2) 中描述。

The short vowel in RP and GA speakers’ pronunciation of the word bet is a front unrounded vowel, whose height is somewhere between cardinal vowels nos. 2 and 3. For most RP and GA speakers, it is closer to cardinal vowel no. 3 than to cardinal vowel no. 2 in height; it is also somewhat more centralized than cardinal vowel no. 3. For convenience’ sake, we will transcribe it as [ε]; its location is depicted in (2) below.

RP 说话者对单词pot的发音中的短元音是后圆唇元音，发音时舌头高度介于 low 和 lowmid 之间（即介于 5 号和 6 号基数元音之间）。它被转录为[ɒ]；它的位置在下面的 (2) 中给出。GA 系统中没有这个元音：GA 说话者在诸如pot之类的词中有元音 [ɑ] 。[ɑ] 是短后圆唇低元音。

The short vowel in the RP speaker’s pronunciation of the word pot is a back rounded vowel which is articulated with a tongue height somewhere between low and lowmid (i.e. between cardinal vowels nos. 5 and 6). It is transcribed as [ɒ]; its location is given in (2) below. This vowel is absent from the GA system: GA speakers have the vowel [ɑ] in words such as pot. [ɑ] is a short back rounded low vowel.

(2) RP 和 GA 短元音

我们使用了pit、pet、pat、pot、putt和put来说明这些元音，因为这些词的发音仅在元音方面有所不同。在讨论元音时，我们还将采用 Wells（1982 年；参见推荐的进一步阅读）采用的词汇集。这些是 Wells 选择的关键词，用来说明 RP 和 GA 之间的相同点和不同点。因此，我们有时会将pit等词中的元音称为 KIT 元音。pet等单词中的元音我们称之为DRESS元音；像pat这样的词有 TRAP 元音；锅之类的话有很多元音；put之类的词有 FOOT 元音，而putt之类的词有 STRUT 元音。

We have used the words pit, pet, pat, pot, putt and put to illustrate these vowels, since these words differ in pronunciation only with respect to the vowel. In discussing vowels, we will also adopt the lexical sets adopted by Wells (1982; see Suggested Further Reading). These are key words selected by Wells to bring out the similarities and differences between RP and GA. We will therefore, at times, refer to the vowel in words such as pit as the KIT vowel. The vowel in words such as pet we will call the DRESS vowel; words such as pat have the TRAP vowel; words such as pot have the LOT vowel; words such as put have the FOOT vowel, and words such as putt have the STRUT vowel.

还有一个元音，如上所述，我们必须在这个阶段考虑。它是第一个元音，出现在大多数说话者对单词about的发音中。这个元音被称为schwa；它是在没有唇的情况下生产的四舍五入，舌体位于元音空间的最中心部分，介于高中音和中低音之间，在后和前之间。Schwa 被转录为 [ə]。这个元音通常比我们刚刚描述的短元音更短，它与那些可能永远不会出现在重读音节中的短元音不同（在about中，它出现在非重读的第一个音节中；在elephant中，它出现在非重读的第一个音节中）第二个音节；在Belinda中，它出现在非重读的首音节和末尾音节中）。这个元音出现在几乎所有说英语的人的讲话中；在后面的章节中，我们将更详细地考虑它与英语重读元音的关系。

There is one further vowel sound, indicated above, which we must consider at this stage. It is the first vowel sound which occurs in most speakers’ pronunciation of the word about. This vowel is referred to as schwa; it is produced without lip rounding, and with the body of the tongue lying in the most central part of the vowel space, between high-mid and low-mid, and between back and front. Schwa is transcribed as [ə]. This vowel is typically even shorter than the short vowels we have just described, and it differs from those in that it may never occur in a stressed syllable (in about, it occurs in the unstressed first syllable; in elephant, it occurs in the unstressed second syllable; in Belinda, it occurs in the unstressed initial and final syllables). This vowel occurs in the speech of almost every speaker of English; in later chapters, we will consider its relation to English stressed vowels in more detail.

练习

Exercises

1 描述在发出下列元音时舌体和嘴唇的位置：

[i]（第一基数元音）

[u]（第 8 基数元音）

[ɑ]（第 5 基数元音）

2 为下列每个词中的元音给出一个适当的元音符号，就像你会说的那样。说 (a) 元音是否圆润，(b) 元音有多靠前或靠前，以及 (c) 有多低或多高（相对于基数元音而言）：

pit apt stock bet put 推杆

笔记。如果您在辅导小组中讨论这些练习，您可能已经开始注意到小组成员的语音差异，这取决于他们所说的口音。显然，如果一个人有西约克郡或纽约市的口音，那么将这些词当作RP 或 GA 说话者来抄写是没有意义的。你应该做的是尝试弄清楚（最好在导师的帮助下）每个元音的质量是什么，并为那个元音采用合适的音标，然后你可以在你的转录中一致地使用它。在适当的时候，我们将更详细地检查口音变化。

Note. If you are discussing these exercises in a tutorial group, you may well already have begun to notice differences in the speech of the members of the group, depending on the accents they speak. Clearly, there is little point, if one has, say, a West Yorkshire or a New York City accent, in transcribing these words as if one were an RP or a GA speaker. What you should do is to try to work out (preferably with the help of a tutor) what the quality of each vowel is, and to adopt an appropriate phonetic symbol for that vowel, which you can then use consistently in your transcriptions. In due course, we will be examining accent variation in more detail.

gueqrecard_001

3在www.wiley.com/go/carrphonetics上收听音轨 3.1。为您听到的每个单词提供尽可能多的语音细节的音标：

大象悸动可疑不平衡鼓励

4个

4 英语语音：元音 (ii)

English Phonetics: Vowels (ii)

4.1 RP 和 GA 长元音

4.1 RP and GA Long Vowels

我们注意到put ([ʊ]) 中的 RP/GA 元音比学校的短；我们也说过比学校里少背少高。我们将在学校将元音转录为 [u:]，其中 ':' 变音符号表示元音长度。这是一个高后圆唇元音，比 [ʊ] 更接近基数元音 8。

We noted that the RP/GA vowel in put ([ʊ]) is shorter than that in school; we also said that it is less back and less high than that in school. We will transcribe the vowel in school as [u:], where the ‘ : ’ diacritic denotes vowel length. This is a high back rounded vowel, articulated closer to cardinal vowel 8 than [ʊ].

我们在第 3 章中描述的fit中的 RP/GA 短元音 [ɪ]是一个相当高、相当靠前的不圆唇元音。它与feet中的 RP/GA 元音不同，后者更长、更靠前、更高。我们将其转录为 [i:]；它是一个高前不圆唇元音，比 [ɪ] 更接近基数元音 1。

The RP/GA short vowel [ɪ], as in fit, which we described in chapter 3, is a fairly high, fairly front, unrounded vowel. It differs from the RP/GA vowel in feet, which is longer, more front and higher. We will transcribe this as [i:]; it is a high front unrounded vowel which is closer to cardinal vowel 1 than [ɪ].

值得注意的是，在 RP 和 GA 中，当to和(s)he等词单独发声时，它们分别包含元音 [u:] 和 [i:]，因此to的发音相同方式如二也。但是像to和(s)he这样的“功能”词（它们不是名词、形容词或动词）通常在没有重读的情况下发出，在这种情况下，它们可以用 schwa ([ə]) 或缩写形式发出，如to eat（发音为 [təi:t] 或 [tui:t]）和she wear（发音为 [ʃəwɔ] 或 [ʃiwɔ:]）。[i:] 的缩写形式也出现在各种后缀中，例如witty中的后缀: [wɪti] 和后缀 in quickly : [kʰwɪkli]。它也出现在单词的非重读音节中，例如pretty : [pʰɹɪti]。^1个

It is worth noting that in RP and GA, when words such as to and (s)he are uttered in isolation, they contain, respectively, the vowels [u:] and [i:], so that to is pronounced in the same way as two and too. But ‘function’ words like to and (s)he (which are not nouns, adjectives or verbs) are often uttered without stress, in which case they may be uttered with a schwa ([ə]), or in a shortened form, as in to eat (pronounced either as [təi:t] or as [tui:t]) and she wore (pronounced either [ʃəwɔ] or [ʃiwɔ:]). The shortened form of [i:] is also found in various suffixes, as in the suffix in witty: [wɪti] and in the suffix in quickly: [kʰwɪkli]. It occurs too in the unstressed syllable of words such as pretty: [pʰɹɪti].1

port和caught的RP 元音比pot和cot长；它是一个中低后圆唇元音，发音比 RP pot and cot中的 [ɒ] 更接近基数元音 6 。我们将其转录为 [ɔ:]。这也是 GA 说话者在像caught这样的词中说出的元音（尽管 GA 元音比 RP 元音稍短）。因此，虽然 GA 和 RP 说话者都区分cot和caught等对，但 GA 在cot中有 [ɑ]而 RP 说话者有 [ɒ]。在 GA 中，单词例如horse和port，元音后有 [ɹ]，通常发为 [hɔɹs] 和（见下文 RP 和 GA 中的 /oʊ/）。

The RP vowel in port and caught is longer than that in pot and cot; it is a low-mid back rounded vowel, articulated closer to cardinal vowel 6 than is the [ɒ] in RP pot and cot. We will transcribe it as [ɔ:]. This is also the vowel which GA speakers utter in words like caught (although the GA vowel is somewhat shorter than the RP vowel). Thus, although both GA and RP speakers distinguish between pairs such as cot and caught, GA has [ɑ] in cot whereas RP speakers have [ɒ]. In GA, words such as horse and port, with an [ɹ] after the vowel, are typically uttered as [hɔɹs] and (see below on /oʊ/ in RP and GA).

RP 和 GA 的短元音 [æ]，如ant中的，正如我们所见，是一个相当低的、相当靠前的、不圆唇的元音。与 aunt不同的是aunt是下背不圆唇元音，发音在第 5 基元音区。aunt 的 RP/GA 元音也比ant的 RP/GA 元音长。我们将其转录为 [ɑ:]。因此，尽管 RP 在 [ɒ]、[ɑ:] 和 [ɔ:] 之间具有三向区分，但 GA 在 [ɑ] 和 [ɔ:] 之间只有双向区分。我们将回到下面重音之间的这种差异。

The RP and GA short vowel [æ], as in ant, is, as we have seen, a fairly low, rather front, unrounded vowel. It differs from that in aunt, which is a low back unrounded vowel, articulated in the region of cardinal vowel 5. The RP/GA vowel in aunt is also longer than the RP/GA vowel in ant. We will transcribe it as [ɑ:]. Thus, whereas RP has a three-way distinction between [ɒ], [ɑ:] and [ɔ:], GA has only a two-way distinction between [ɑ] and [ɔ:]. We will return to this difference between the accents below.

说 RP 和 GA 的人会在bird、heard、dearth等词中发出长元音，尽管说 GA 的人会在这些词中发出 [ɹ]，而说 RP 的人不会。这个元音的发音与 schwa 的发音几乎相同：它在高/低和前/后维度上都处于中心位置，并且不圆润。与 schwa 不同，它出现在重读音节中。我们将其转录为 [ɜ:]。

RP and GA speakers utter a long vowel in words like bird, heard, dearth, although GA speakers utter an [ɹ] in words such as these, while RP speakers do not. The articulation for this vowel is pretty much the same as that for schwa: it is central on both the high/low and front/back dimensions, and is unrounded. Unlike schwa, it appears in stressed syllables. We will transcribe it as [ɜ:].

我们可以将这些元音在元音空间中的大致发音区域描述如下：

We may depict the approximate areas of articulation of these vowels in the vowel space as follows:

(1) RP和GA长元音

我们将按照 Wells（1982 年；参见推荐的进一步阅读），将 [i:] 称为 FLEECE 元音，将 [u:] 称为 GOOSE 元音，将 [ɜ:] 称为 NURSE 元音。Wells 使用三个关键字来表示 [ɔ:] 元音：THOUGHT、FORCE 和 NORTH；我们将在稍后阶段了解原因。类似地，Wells 对 [ɑ:] 元音使用了三个关键字：START、BATH 和 PALM。原因之一是在集合 BATH 中，GA 有 [æ]，而 RP 有 [ɑ:]，而在集合 START 和 PALM 中，GA 和 RP 都有 [ɑ:]。

We will, following Wells (1982; see Suggested Further Reading), refer to [i:] as the FLEECE vowel, [u:] as the GOOSE vowel and [ɜ:] as the NURSE vowel. Wells uses three key words for the [ɔ:] vowel: THOUGHT, FORCE and NORTH; we will see why at a later stage. Similarly, Wells uses three key words for the [ɑ:] vowel: START, BATH and PALM. One of the reasons for this is that in words of the set BATH, GA has [æ], whereas RP has [ɑ:], whereas in words of the sets START and PALM, both GA and RP have [ɑ:].

4.2 RP 和 GA 双元音

4.2 RP and GA Diphthongs

到目前为止，在我们考虑的所有 RP 和 GA 元音中，发音器在元音发音的整个过程中或多或少地保持在相同的位置。这意味着元音质量（元音发音过程中产生的声学效果）或多或少保持不变。那种元音是单元音。然而，有些元音并非如此。这种元音，称为双元音，在其产生过程中需要改变发音器的位置，从而改变产生的元音质量。双元音是一个元音，其音质在一个音节内发生变化。双元音不仅仅是两个元音的序列。例如，在单词seeing的 RP 和 GA 发音中([si:ɪŋ])，元音 [i:] 后跟元音 [ɪ]，但结果序列不是双元音，因为 [i:] 和 [ɪ] 不在同一个音节中：seeing有两个音节，第一个以 [i:] 结尾，第二个以 [ɪ] 开头。

In all of the RP and GA vowel sounds we have considered thus far, the articulators remain more or less in the same position throughout the articulation of the vowel. This means that the vowel quality (the acoustic effect created during the articulation of the vowel) remains more or less constant. That kind of vowel is a monophthong. However, there are vowel sounds in which this is not the case. This kind of vowel sound, called a diphthong, entails some kind of change of position of the articulators during its production, and thus a change in the vowel quality produced. A diphthong is a vowel whose quality changes within a syllable. A diphthong is not simply a sequence of two vowels. For instance, in both the RP and the GA pronunciations of the word seeing ([si:ɪŋ]), the vowel [i:] is followed by the vowel [ɪ], but the resulting sequence is not a diphthong, because the [i:] and the [ɪ] are not in the same syllable: seeing has two syllables, the first of which ends in [i:] and the second of which begins with [ɪ].

让我们从以类似 [ɪ] 的音质结尾的双元音开始。在诸如sigh、rye、bide等词的 RP 和 GA 发音中，元音以类似 [a] 的音质开始（在基数元音 4 区域）并以类似 [ɪ] 的音质结束。我们将其转录为 [aɪ]。

Let us begin with diphthongs which end in an [ɪ]-like quality. In the RP and GA pronunciations of words such as sigh, rye, bide, etc., the vowel begins with an [a]-like quality (in the region of cardinal vowel 4) and ends in an [ɪ]-like quality. We will transcribe this as [aɪ].

在say、ray、bayed等的 RP 和 GA 发音中，元音以类似 [e] 的音质开始（在基数元音 2 区域）并以类似 [ɪ] 的音质结束。我们将其转录为 [eɪ]。在诸如hair之类的词中，GA 发音是单元音 [e]，后跟 [ɹ]。

In the RP and GA pronunciations of say, ray, bayed, etc., the vowel begins with an [e]-like quality (in the region of cardinal vowel 2) and ends in an [ɪ]-like quality. We will transcribe this as [eɪ]. In words such as hair, the GA pronunciation is a monophthongal [e], followed by an [ɹ].

在soy、Roy、buoyed等的 RP 和 GA 发音中，元音以类似 [ɔ] 的音质开始（在基数元音 6 区域）并以类似 [ɪ] 的音质结束。我们将其转录为 [ɔɪ]。

In the RP and GA pronunciations of soy, Roy, buoyed, etc., the vowel begins with an [ɔ]-like quality (in the region of cardinal vowel 6) and ends in an [ɪ]-like quality. We will transcribe this as [ɔɪ].

我们可以在元音空间图中表示这些双元音如下：

(2) 以[ɪ]结尾的RP和GA双元音

Wells 使用关键词 FACE 表示 [eɪ] 双元音，CHOICE 表示 [ɔɪ] 元音，PRICE 表示 [aɪ] 元音；当证明有用时，我们将遵循这种做法。

Wells uses the key word FACE for the [eɪ] diphthong, CHOICE for the [ɔɪ] vowel and PRICE for the [aɪ] vowel; we will follow this practice when it proves useful.

RP 和 GA 中有两个双元音以类似 [ʊ] 的音质结尾。其中第一个以低沉的、相当靠后的、不圆润的质量开始。它出现在单词的 RP 和 GA 发音中，例如how, now, loud。我们将这个双元音转录为[aʊ]。

There are two diphthongs in RP and GA which end in an [ʊ]-like quality. The first of these begins with a low, rather back, unrounded quality. It is found in the RP and GA pronunciations of words such as how, now, loud. We will transcribe this diphthong as [aʊ].

这些双元音中的第二个开始，在 GA 说话者和更保守的 RP 说话者中，具有类似 [o] 的品质。它出现在sew、roe、toad等词中。我们将其转录为 [oʊ]。在更现代的 RP 使用者中，诸如此类的词的发音具有类似 [ɔʊ] 的音质。²诸如sport之类的词，在 RP 中以长元音 [ɔ:] 发声但没有 [ɹ]，在 GA 中以 [o] 后跟 [ɹ] 发声。

The second of these diphthongs begins, among GA speakers, and among more conservative RP speakers, with an [o]-like quality. It occurs in words such as sew, roe, toad. We will transcribe this as [oʊ]. Among more modern RP speakers, words such as these are pronounced with an [ɔʊ]-like quality.2 Words such as sport, uttered with the long vowel [ɔ:] but no [ɹ] in RP, are uttered with an [o] followed by an [ɹ] in GA.

这两个双元音可以在元音空间内表示如下：

These two diphthongs may be represented within the vowel space as follows:

(3) 以[ʊ]结尾的RP和GA双元音

对于这些双元音，Wells 使用关键字 MOUTH 表示 [aʊ] 元音，使用 GOAT 表示 [oʊ] 双元音。

For these diphthongs, Wells uses the key word MOUTH for the [aʊ] vowel and GOAT for the [oʊ] diphthong.

许多 RP 说话者发出一系列双元音，这些双元音以类似 [ə] 的音质结尾，即 schwa。由于 schwa 在元音空间的中心发音，这些通常被称为居中双元音。这些双元音中的第一个以类似 [ɪ] 的音质开头。它出现在诸如here和pier之类的词中。我们将其转录为 [ɪə]。

Many RP speakers utter a series of diphthongs which end in an [ə]-like quality, i.e. schwa. Since schwa is pronounced in the centre of the vowel space, these are often called centring diphthongs. The first of these diphthongs begins with an [ɪ]-like quality. It occurs in words such as here and pier. We will transcribe this as [ɪə].

另一个此类双元音以类似 [ε] 的性质开始（在基数元音 3 区域）。这出现在hair和pear等词的 RP 发音中。我们将其转录为 [εə]。一些说 RP 的人用 [ε:] 来发音这类单词，这是一个根本不是双元音的长元音，但更像是基数元音 3 区域中元音的长版本。

Another diphthong of this sort begins with an [ε]-like quality (in the region of cardinal vowel 3). This occurs in the RP pronunciation of words such as hair and pear. We will transcribe this as [εə]. Some RP speakers pronounce words of this sort with [ε:], a long vowel which is not a diphthong at all, but is more like a long version of a vowel in the region of cardinal vowel 3.

第三个这样的双元音以类似 [ʊ] 的音质开头，出现在诸如tour和pure之类的词中。我们将其转录为 [ʊə]。一些说 RP 的人将其中一些词（例如moor）发音为基数元音 6 区域的长单元音。如果遇到这种情况，您可以合理地将其转录为 [ɔ:]。

A third such diphthong begins with an [ʊ]-like quality, and occurs in words such as tour and pure. We will transcribe this as [ʊə]. Some RP speakers pronounce some of these words (e.g. moor) as a long monophthong in the region of cardinal vowel 6. If you encounter this, you may reasonably transcribe it as [ɔ:].

这三个双元音可以表示如下：

These three diphthongs may be represented as follows:

(4) 以 [ə] 结尾的 RP 双元音（居中双元音）

这些被称为居中双元音，因为 schwa 位于元音空间的中心。Wells 使用关键词 NEAR 表示 [ɪə] 元音，CURE 表示 [ʊə] 元音，SQUARE 表示 [εə] 元音。这些双元音在 GA 中都没有。它们在 RP 中的存在是由于在 RP 的历史发展中元音后 [ɹ] 的丢失：schwa 可以说是 [ɹ] 的唯一残留痕迹，它曾经存在于 RP 演变而来的口音中，here、hair和pure等词的发音, 在 GA 中发音为 [hi:ɹ]、[heɹ] 和 [pʰu:ɹ]。在 RP 中，通常会发现单元音变体。对于现代 RP 中的 SQUARE 元音，常见的是长单元音：[ε:]。对于 CURE 元音，该词组中的许多单词现在由 RP 说话者用长单元音 [ɔ:] 发音，如单词sure发音为 [ʃɔ:]。

These are called centring diphthongs since schwa is located at the centre of the vowel space. Wells uses the key words NEAR for the [ɪə] vowel, CURE for the [ʊə] vowel, and SQUARE for the [εə] vowel. These diphthongs are all absent in GA. Their presence in RP results from the loss of [ɹ] after vowels in the historical development of RP: the schwa is, as it were, the only remaining trace of the [ɹ] which once existed in the accents from which RP evolved, in the pronunciation of words such as here, hair and pure, which are pronounced [hi:ɹ], [heɹ] and [pʰu:ɹ] in GA. In RP, it is common to find a monophthongal variant. For the SQUARE vowel in contemporary RP, it is common to find a long monophthong: [ε:]. For the CURE vowel, many words of that lexical set are now pronounced by RP speakers with the long monophthong [ɔ:], as in the word sure, pronounced [ʃɔ:].

笔记

Notes

1在quickly、witty和pretty等词中，英语的口音因词尾元音而异。有些有 [i] 型元音，而另一些有 [ɪ] 型元音。前一个元音有时被认为比后者“更紧”，这些词中带有 [i] 的重音有时被描述为具有“ HAPPY tensing”。后一术语归功于 Wells（1982 年；请参阅推荐的进一步阅读）。

1 Accents of English vary with respect to the final vowel in words such as quickly, witty and pretty. Some have an [i]-type vowel, while others have an [ɪ]-type vowel. The former vowel is sometimes said to be ‘more tense’ than the latter, and accents with [i] in these words are sometimes described as having ‘HAPPY tensing’. The latter term is due to Wells (1982; see Suggested Further Reading).

2在较年轻的 RP 使用者中，这些双元音经常与前置不圆唇第二元素一起发音；我们可以将这些发音转录为 [ə ı ] 和 [a ı ]，其中第二个符号表示相对较高的中央不圆唇元音，例如coke ： [kʰə ı k] 和以下：[da ı n]。其效果是让可乐听起来更像蛋糕，喝起来更像正餐。还有许多其他这样的对；关于它们的主要观点是，这对仍然是不同的，但不像过去那样明显。

2 Among younger speakers of RP, these diphthongs are frequently uttered with a fronted unrounded second element; we could transcribe these pronunciations as [əı] and [aı], where the second symbol denotes a relatively high central unrounded vowel, as in coke: [kʰəık] and down: [daın]. The effect is to make coke sound rather like cake, and down rather like dine. There are many other such pairs; the principal point about them is that the pairs are still distinct, but less markedly so than in the past.

练习

Exercises

1 按照语音抄写你在以下每个单词中发出的元音：

caught court cot blew put dearth death feel fill 山姆圣歌

（见第 3 章练习 2 下的注释。）

2 对于母语为英语的人：以下单词中的元音通常是 RP 中的双元音。对于每个单词，按照您通常所说的那样，用适当的符号语音转录元音。如果它是您演讲中的双元音，请描述初始和最终的嘴唇和舌头配置。如果它不是双元音，说明它有多靠前/靠前，有多高/多低，是否圆润。

恐惧公平旅游迟到叹息侧加入蟾蜍拖磅

（再次参见第 3 章练习 2 下的注释。）

gueqrecard_001

3在www.wiley.com/go/carrphonetics收听曲目 4.1。尽可能详细地转录您听到的每个单词：

推车结束神圣的神性平静安详

5个

5 音位原则

The Phonemic Principle

5.1 简介：语言知识

5.1 Introduction: Linguistic Knowledge

到目前为止，我们一直在处理语音学，也就是说（正如我们所定义的那样），研究人类语音（尽管我们只处理了英语语音学，特别是专门的发音语音学，忽略了有关的重要事实我们一直在讨论的语音的声学特性）。从今以后，我们将处理语音学和语音学。我们会声称，音位学所涉及的不仅仅是人类语音本身的特性. 语音学是对某些类型的心理组织的研究。特别是，它研究某些类型的心理范畴、心理存储的表征以及关于这些范畴和表征的概括。按照这种观点，音韵学本身并不是对人类语音的研究，尽管语音学和音韵学密不可分。本章的重点是展示两者之间的区别，并开始向读者介绍英语的音韵学。让我们首先考虑一些关于什么是了解一门语言的一般性问题。

We have been dealing, thus far, with phonetics, that is (as we have defined it), with the study of human speech sounds (although we have dealt exclusively with English phonetics, and in particular, exclusively articulatory phonetics, ignoring important facts about the acoustic properties of the speech sounds we have been discussing). We will, henceforth, be dealing with phonology, as well as phonetics. Phonology, we will claim, is to do with something more than properties of human speech sounds per se. Phonology is the study of certain sorts of mental organization. In particular, it is the study of certain types of mental category, mentally stored representations, and generalizations concerning those categories and representations. On this view, phonology is not the study of human speech sounds per se, although phonetics and phonology are inextricably intertwined. The point of this chapter is to demonstrate what the difference between the two is, and to begin to introduce the reader to the phonology of English. Let us begin by considering some general questions concerning what it is to know a language.

让我们假设，当我们说某人懂一种语言时，从以该语言为母语的意义上说，他或她处于某种精神状态，或者拥有某种语言知识。显然，对母语的了解在很大程度上是无意识的知识。它似乎包含语义知识（与词、词组和句子的意义有关）和句法知识（与词的句法类别、词组和句子的结构以及词、词组和从句之间的句法关系有关） ). 我们知道是这样，因为说话者能够根据这些知识做出句法和语义判断。例如，母语为English 可以判断Who did you see Graham with? 是英文句子，who did you see Graham and? 不是。说话者再次凭直觉知道，两者之间的区别不仅仅是单词的存在与单词的存在之间的区别。他或她还凭直觉（不一定是完全有意识地）知道在什么意义上他告诉那个人他知道的人是模棱两可的，以及在什么意义上对该词序列的两种解释在结构和含义上不同于他告诉那个人他是如何知道的，除了一个序列包含谁和另一个的表面事实之外怎么样。这种知识显然是无意识的知识，因为我们不需要任何指导就能做出这样的判断，而且我们可以在没有任何关于英语句法和语义的有意识知识的情况下做出这些判断（即使一个人有对名词或动词可能是什么，或者with、and、who和how可能是什么的句法类别一无所知）。

Let us assume that when we say that someone knows a language, in the sense of being a native speaker of that language, he or she is in a certain mental state, or possesses a certain sort of linguistic knowledge. Knowledge of a native language is, apparently, largely unconscious knowledge. It appears to contain semantic knowledge (to do with the meanings of words, phrases and sentences) and syntactic knowledge (to do with the syntactic categories of words, with the structure of phrases and sentences and with the syntactic relations between words, phrases and clauses). We know that this is so, since speakers are able to make syntactic and semantic judgements, based on that knowledge. For instance, a native speaker of English can judge that Who did you see Graham with? is an English sentence, and that Who did you see Graham and? is not. The speaker knows, again intuitively, that the difference between the two amounts to more than the difference between the mere presence of the word and as opposed to the presence of the word with. He or she also knows intuitively (not necessarily fully consciously) in what sense He told the man who he knew is ambiguous, and in what sense the two interpretations of that sequence of words differs in structure and meaning from He told the man how he knew, over and above the superficial fact that one sequence contains who and the other how. That knowledge is clearly unconscious knowledge, since we require no instruction to be able to make such judgements, and we can make them in the absence of any conscious knowledge whatsoever of the syntax and semantics of English (one could make such judgements even if one had not the faintest idea of what a noun or a verb might be, or what the syntactic categories of with, and, who and how might be).

我们将在本书中采取这样的观点，即说话者对其母语的（大部分）无意识知识也必须包含语音知识。许多语言学家持这种观点的原因之一是，说话者可以做出判断，据称，这些判断在某种意义上与对句法事态所做的判断是平行的。例如，以英语为母语的人可以说出一个单词中有多少个音节，而无需有意识地知道音节是什么。这表明母语人士具有识别音节的能力，即使对音节的识别处于意识水平以下。据称，以类似的方式，以英语为母语的人可以分辨出片段序列 [blʌg]，被认为是一个单词的发音，是一个英语序列，而片段的序列 [tʰlʌg] 则不是，尽管事实上她或他可能从未在她或他的生活中听过任何序列。让我们假设，在做出这样的判断时，以英语为母语的人获得了一种无意识的知识，这种知识构成了“英语的音系学”。

We will take the view in this book that a speaker’s (largely) unconscious knowledge of his or her native language(s) must also contain phonological knowledge. One of the reasons many linguists take this view is that speakers can make judgements which, it is claimed, are in some sense parallel to those made with respect to syntactic states of affairs. For instance, a native speaker of English can tell how many syllables there are in a word without having the faintest idea, consciously, as to what a syllable is. This shows that the native speaker has the ability to recognize syllables, even if the recognition of syllables lies below the level of consciousness. In a similar fashion, it is claimed, a native speaker of English can tell that the sequence of segments [blʌg], considered as an utterance of a word, is an English sequence, whereas the sequence of segments [tʰlʌg] is not, despite the fact that she or he may well never have heard either sequence in her or his life. Let us postulate that, in making such judgements, the native speaker of English gains access to a kind of unconscious knowledge which constitutes ‘the phonology of English’.

在本书中，我们的任务将是开始以一种基本的方式考虑这种知识的形式。在这种观点下，音韵学不同于语音学，因为它研究的不是语音本身，而是心理能力和很大程度上是无意识的心理状态。显然，音韵学家必须密切关注语音及其特性；它们将构成语音学家用来支持他或她关于说话者无意识语音知识的假设的大部分证据，但它们并不构成他或她的探究对象本身。

Our task, in this book, will be to begin to consider, in an elementary way, what form that knowledge takes. The discipline of phonology, under this view, differs from that of phonetics, since it is the study, not of speech sounds per se, but of mental abilities and largely unconscious mental states. Clearly, the phonologist must pay close attention to speech sounds and their properties; they will constitute much of the evidence the phonologist brings to bear on his or her hypotheses about speakers’ unconscious phonological knowledge, but they do not constitute his or her object of inquiry as such.

5.2 对比与可预测性：音素

5.2 Contrast vs Predictability: The Phoneme

让我们首先考虑英语和韩语中的清不送气塞音和清送气塞音。大多数英语口音的使用者习惯性地发出送气和不送气的清塞音。下面的英语数据展示了这两种情况。^1个

Let us begin by considering voiceless unaspirated and voiceless aspirated stops in English and Korean. Speakers of most accents of English habitually utter both aspirated and unaspirated voiceless stops. The following English data exhibit both of these.1

(1) 英语中的送气和不送气清塞音

（一种）	[ˈpʰu:ɫ]	水池
（乙）	[əˈpʰɪə]	出现
（C）	[ˈspɜ:t]	喷
(四)	[dəˈspaɪt]	尽管
(五)	[ˈtʰɒp]	最佳
（F）	[əˈtʰæk]	攻击
（G）	[ˈstɒp]	停止
（H）	[dəstɹɔɪ]	破坏
（一世）	[ˈkʰɪlɪŋ]	杀戮
(十)	[əˈkʰɹu:]	应计
(k)	[ˈskoʊɫd]	骂
(l)	[dɪˈskʌvə]	发现

这些数据中某些符号之前的变音符号（[ˈpʰu:ɫ] 中“p”符号之前的变音符号表示重读音节的开头。

The diacritic which precedes certain symbols in these data (the one which precedes the ‘p’ symbol in [ˈpʰu:ɫ] indicates the beginning of a stressed syllable.

从这些数据来看，清塞音在重读音节的开头时似乎是送气的，如在pool和appear 中，但在清音齿龈擦音之前是不送气的，如在spurt中。也就是说，在这些数据中，只要出现不送气的清音塞，送气的就不会出现，反之亦然。将英文数据与以下韩文数据进行比较：

From these data, it appears that voiceless stops are aspirated when they are at the beginning of a stressed syllable, as in pool and appear, but unaspirated when preceded by a voiceless alveolar fricative, as in spurt. That is, in these data, wherever the unaspirated voiceless stops appear, the aspirated ones do not, and vice versa. Compare the English data with the following data from Korean:

(2) 韩语中的送气和不送气清塞音

（一种）	[pʰul]	'草'
（乙）	[脉冲]	'火'
（C）	[tʰal]	'面具'
(四)	[总计]	'月亮'
(五)	[kʰεda]	'挖'
（F）	[kεda]	'折叠'

在这些韩语数据中，送气和不送气清塞音可能出现在同一个地方（单词的开头）。一个单词中可能出现给定声音的位置范围称为它的分布。在我们查看的英语数据中，不送气和送气停止的分布是相互排斥的：在你得到一种停止的地方，你永远不会得到另一种。这称为互补分布。

In these Korean data, aspirated and unaspirated voiceless stops may occur in the same place (at the beginning of a word). The range of places within a word which a given sound may occur in is called its distribution. In the English data we have looked at, the distribution of unaspirated and aspirated stops is mutually exclusive: where you get one kind of stop, you never get the other. This is called complementary distribution.

此外，如果我们采用英语数据中的塞音 [t] 和 [tʰ]，很明显它们在语音上是相似的：都是塞音，都是清音，都是齿槽音。然而，对于大多数说英语的人来说，肺泡停在，比方说，still和till听起来是一样的，尽管事实上前者是不送气的，而后者是吸气的。对于说英语的人来说，这两个语音上不同的声音“算作同一件事”。我们不能毫无矛盾地说它们同时是“相同的声音”和“不同的声音”。我们要说的是，虽然它们在语音上不同，但它们在语音上是等价的. 也就是说，两种类型的停止对应于，被解释为属于一个单一的心理类别。我们将把这样的类别称为音素。说英语的人仅根据三个音素来解释 [p]、[pʰ]、[t]、[tʰ]、[k] 和 [kʰ] 六个音段：/p/、/t/ 和 /k/。我们可以这样描述：

Furthermore, if we take, say, the stops [t] and [tʰ] in the English data, it is clear that they are phonetically similar: both are stops, both are voiceless, both are alveolar. And yet, for most speakers of English, the alveolar stops in, say, still and till sound the same, despite the fact that the former is unaspirated and the latter aspirated. For the English speaker, these two phonetically distinct sounds ‘count as the same thing’. We cannot say, without contradiction, that they are simultaneously ‘the same sound’ and ‘not the same sound’. What we will say is that, while they are phonetically distinct, they are phonologically equivalent. That is, the two types of stop correspond to, are interpreted as belonging to, a single mental category. We will refer to such a category as a phoneme. The English speaker interprets the six phonetic segments [p], [pʰ], [t], [tʰ], [k] and [kʰ] in terms of only three phonemes: /p/, /t/ and /k/. We may depict this as follows:

(3) 英语清音塞音

这里最上面的一行代表三种清音停止音素（心理类别），根据这些音素感知六种类型的语音片段。音素与其相关语音段之间的关系是一种实现, 例如，音位 /p/ 在清齿龈擦音后被实现为 [p]，而在其他地方被实现为 [pʰ]。最重要的一点是，根据我们目前所看到的数据，英语中是否有抱负是完全可以预测的：对于清音停止 will 和 will 的语境，有一个概括，可以表达为一般规则不会被吸入。对于大多数英语口音，这种概括是儿童在将英语作为母语时内化的。这种概括构成了以母语为母语的人在了解他们的母语时所知道的一部分，即使这种知识在很大程度上是无意识的知识。完全可以从上下文预测的音素的实现称为它的音位变体. 因此我们说 [p] 和 [pʰ] 是大多数英语口音中 /p/ 音素的同位异音。我们声称，以英语为母语的人拥有音位（这是心理范畴）和语音概括或规则，作为他们（很大程度上是无意识的）母语知识的一部分，并且母语者根据这些范畴感知他们听到的变位音，概括。

The top line here represents the three voiceless stop phonemes (mental categories) in terms of which the six types of phonetic segment are perceived. The relationship between phonemes and their associated phonetic segments is one of realization, so that the phoneme /p/, for instance, is realized as [p] after a voiceless alveolar fricative, and as [pʰ] elsewhere. The most important point is that, on the data we have seen thus far, aspiration or the lack of it is entirely predictable in English: there is a generalization, expressible as a general rule, as to the contexts in which voiceless stops will and will not be aspirated. For most accents of English, this generalization is one that is internalized by children when they acquire English as their native language. The generalization forms part of what native speakers know in knowing their native language, even if that knowledge is largely unconscious knowledge. Realizations of a phoneme which are entirely predictable from context are called its allophones. We therefore say that [p] and [pʰ] are allophones of the /p/ phoneme in most accents of English. We are claiming that native speakers of English possess phonemes (which are mental categories) and phonological generalizations or rules as part of their (largely unconscious) knowledge of their native language, and that native speakers perceive the allophones they hear in terms of those categories and generalizations.

将英语的情况与韩语的情况进行比较。很明显，韩语中送气和不送气清塞音的分布是重叠的：至少有一个地方（在单词的开头）可能出现任何一种声音。这种分布称为平行分布，其中“平行”表示“在某种程度上重叠”。

Compare the English situation with the Korean one. It is clear that the distribution of aspirated and unaspirated voiceless stops in Korean is overlapping: there is at least one place (at the beginning of words) in which either type of sound may occur. This kind of distribution is referred to as parallel distribution, where ‘parallel’ means ‘overlapping to some degree’.

此外，送气塞音和不送气清塞音之间的区别在韩语中会产生至关重要的差异：当说韩语的人说 [pʰul] 时，它与 [pul] 的意思不同。这两种发音之间的差异被认为是语义对比。只有一个声音不同的词对称为最小词对。它们的存在很重要，因为它们表明所讨论的两个声音都是平行分布的并且在语义上是对比的。

Furthermore, the distinction between aspirated and unaspirated voiceless stops can make a crucial difference in Korean: when the Korean speaker says [pʰul], it does not mean the same thing as [pul]. The difference between the two sounds is said to be semantically contrastive. Pairs of words which differ with respect to only one sound are called minimal pairs. Their existence is important, since they demonstrate that the two sounds in question are both in parallel distribution and semantically contrastive.

因此，我们想说的是，与说英语的人不同，韩国人根据六种不同的方式感知六个送气和不送气的清塞音 [p]、[pʰ]、[t]、[tʰ]、[k] 和 [kʰ]心理类别。也就是说，例如，[p] 是 /p/ 音素的实现，而 [pʰ] 是不同的 /pʰ/ 音素的实现。我们可以这样描述（部分）²韩国系统：

We therefore want to say that, unlike the English speaker, the Korean perceives the six aspirated and unaspirated voiceless stops [p], [pʰ], [t], [tʰ], [k] and [kʰ] in terms of six different mental categories. That is, [p], for instance, is a realization of the /p/ phoneme, whereas [pʰ] is a realization of a distinct /pʰ/ phoneme. We may depict (part of)2 the Korean system thus:

(4) 部分韩语清音塞音

送气和不送气清塞音之间的区别在韩语中是音位的，但在英语中是同位音的。说英语和说韩语的人习惯性地说送气和不送气的清塞音。因此，在语音层面上，就双唇、牙槽和软腭清塞音而言，这两种语言是等同的。但在音素层面（心理层面），两种语言截然不同：说韩语的人有六个心理类别，而说英语的人只有三个。就清音塞音而言，说韩语的人的音素对比是说英语的人的两倍。因此，说英语的人在学习发音和感知韩语清音塞音时遇到的困难是一种心理问题一; 这是一个语音困难，而不是纯粹的发音困难。

The distinction between aspirated and unaspirated voiceless stops is phonemic in Korean but allophonic in English. Both English and Korean speakers habitually utter both aspirated and unaspirated voiceless stops. On the phonetic level, the two languages are therefore equivalent as far as bilabial, alveolar and velar voiceless stops are concerned. But at the phonemic level (the mental level), the two languages are quite distinct: the Korean speaker has six mental categories where the English speaker has only three. As far as voiceless stops are concerned, Korean speakers have twice as many phonemic contrasts as English speakers. The difficulty which the English speaker encounters in learning to pronounce and perceive Korean voiceless stops is therefore a mental one; it is a phonological difficulty, not a purely articulatory one.

这并不是否认在学习说另一种语言时可能存在纯粹的发音困难（发音不习惯发音的新型声音的困难）。例如，大多数正在学习说英语的日本人都必须学习发 [l] 音，而他们并不习惯发这个音。当外语学习者面临这项任务时，他们通常会发出母语中与目标声音相似的声音：在这种情况下，拍子 [ɾ] 与 [l] 一样，是浊音和齿槽音。同样，试图掌握英语发音 [ð] 的法语使用者经常会发出浊齿龈擦音 [z] 或浊牙塞音 [d̪]，这是她或他在母语中习惯说的. 前者类似于目标音是浊擦音，而后者类似于浊齿音。外语发音的此类问题很普遍。但它们与我们刚才讨论的问题类型不同。

This is not to deny that there can be purely articulatory difficulties in learning to speak another language (difficulties in articulating new types of sound which one is not in the habit of articulating). For instance, most speakers of Japanese who are learning to speak English will have to learn to pronounce the sound [l], which they are not in the habit of pronouncing. When learners of a foreign language face this task, they often utter a sound from their native language which is similar to the target sound: in this case, the tap [ɾ] which, like [l], is voiced and alveolar. Similarly, a speaker of French who is trying to master the English sound [ð] will often utter the voiced alveolar fricative [z] or the voiced dental stop [d̪], which she or he is used to uttering in her or his native language. The former is similar to the target sound in being a voiced fricative, while the latter is similar in being a voiced dental sound. Such problems with the pronunciation of foreign languages are widespread. But they are distinct in kind from the kind of problem we have just discussed.

我们也不必否认，一门外语的发音可能存在困难，这既涉及纯粹的发音困难，也涉及语音困难。例如，正在学习韩语的说英语的人必须学会发音不同于浊塞音、送气清塞音和不送气清塞音的第三种塞音。这些是用“声门张力”发音的韩语清塞音：在它们的产生过程中，声带不振动，但声带也不会分开，就像清塞音一样；相反，声带是收缩的。

We need not deny either that there may be difficulties in the pronunciation of a foreign language which involve both purely articulatory and phonological difficulties. For instance, the English speaker who is learning Korean must learn to articulate a third kind of stop which is distinct from voiced stops, aspirated voiceless stops and unaspirated voiceless stops. These are the voiceless stops of Korean which are articulated with ‘glottal tension’: during their production, the vocal cords do not vibrate, but nor are the vocal cords spread apart, as they are for the voiceless aspirated stops; rather, the vocal cords are constricted.

说英语的人还必须学会（在某种意义上）理解韩语中所有三种停顿之间的区别；由于声门收缩的清塞音是一种新的声音类别，对于说英语的人来说，它们听起来像是他或她更习惯听到的塞音（例如，浊塞音）。这是一个语音困难，加上说英语的人所遇到的纯发音困难也有。然而，从我们在这里看到的数据可以清楚地看出，存在一种完全是语音方面的困难，正是这种困难证明我们有理由对前面章节中讨论的发音语音学进行区分，这构成了对语音本身和语音发音的研究，以及语音学，即对我们解释这些语音的心理范畴系统的研究。

The English speaker must also learn to (in a sense) perceive the distinction between all three sorts of stop in Korean; since the glottally constricted voiceless stops are a new category of sound, they may seem to the English speaker to sound like stops he or she is more used to hearing (voiced stops, for instance). And that is a phonological difficulty, added to the purely articulatory one which the English speaker also has. However, it is clear from the data we have looked at here that there is a type of difficulty which is exclusively phonological, and it is that kind of difficulty which justifies our making a distinction between the kind of articulatory phonetics discussed in the preceding chapters, which constitutes the study of the articulation of speech sounds in and of themselves, and phonology, the study of the system of mental categories in terms of which we interpret those speech sounds.

在检查韩语和英语清塞音之间的语音差异时，我们采用了所谓的音位原则，它由两组两个标准组成，如下所示：

In examining the phonological differences between Korean and English voiceless stops, we have adopted what is known as the phonemic principle, which consists of two sets of two criteria, as follows:

(5) 音位原则

在以下情况下，两个或多个声音是同一音素的实现：

(a) 它们处于互补分布

和

(b) 它们在语音上相似。

在以下情况下，两个或多个声音是不同音素的实现：

(a) 它们是平行（重叠）分布

和

(b) 它们用于表示语义对比。

正是根据音位原则，我们说送气音的语音差异在英语中是同位异音，而在韩语中是音位。

It is on the basis of the phonemic principle that we say that phonetic differences involving aspiration are allophonic in English but phonemic in Korean.

我们刚刚看到一个例子，说韩语的人比说英语的人有更多的音素对比。现在让我们看看另一组数据，情况正好相反。某些苏格兰英语的母语人士习惯性地发出我们表示为“[ɾ]”和“[l]”的语音，即浊音牙槽音和浊音侧牙槽音（如rip和lip）。说韩语的人也是如此。以下是包含这些发音的苏格兰英语和韩语单词的一些示例：

We have just seen a case where the Korean speaker has more phonemic contrasts than the English speaker. Let us now look at another set of data where the converse is the case. Native speakers of some varieties of Scottish English habitually utter the speech sounds we have represented as ‘[ɾ]’ and ‘[l]’, i.e. the voiced alveolar tap and the voiced lateral alveolar approximant (as in rip and lip). So do speakers of Korean. Here are some examples of Scottish English and Korean words which contain those sounds:

(6) 苏格兰英语和韩语中的 [ɾ] 和 [l]

gueq05_020

虽然说苏格兰英语和韩语的人习惯性地发这两个音，但我们可以预测，许多正在学习说这种苏格兰英语的韩语母语者会发现 [l] 和 [ɾ] 之间的区别，当他们说苏格兰英语，比较难掌握。从表面上看，这是令人费解的，因为正如我们刚才所说，说韩语的人可以毫不费力地发出这两个音，而且很可能在开始学习苏格兰英语之前很久就已经发出了数千个音。那么问题出在哪里呢？一种可以立即排除的可能性是说韩语的人正在遇到某种身体上的发音困难的建议：正如我们所看到的那样，显然情况并非如此，这两种声音中的任何一种对他们来说都是新的。

While speakers of Scottish English and Korean habitually utter both sounds, we can predict that many native speakers of Korean who are learning to speak this variety of Scottish English would find the distinction between [l] and [ɾ], when they speak Scottish English, rather difficult to get the hang of. On the face of it, this is puzzling because, as we have just said, Korean speakers have no difficulty in uttering the two sounds, and may well have uttered many thousands of them, long before beginning to learn Scottish English. So wherein does the problem reside? One possibility that can be immediately discounted is the suggestion that Korean speakers are encountering some kind of physical, articulatory difficulty: it is clearly not the case, as we have seen, that either of the sounds is new to them.

困难是精神上的，如果检查上面 (6) 中的数据表，很明显，在苏格兰英语中，这两个声音可能出现在单词中的相同位置，例如单词的开头, 或元音之间。此外，两个单词可能仅在片段 [ɾ] 和 [l] 方面有所不同：涉及这两个声音的成对最少（例如 [ɾæm] 与 [læm]）。在这种苏格兰英语中，[ɾ] 和 [l] 是平行分布的，可以起到表示语义对比的作用。重要的是要记住，当我们说语音差异是对比时，我们指的是语义对比，而不是声音之间的语音差异。

The difficulty is of a mental nature, and if one examines the table of data in (6) above, it is clear that, in Scottish English, the two sounds may occur in the same places within a word, e.g. at the beginning of words, or between vowels. Furthermore, two words may differ solely with respect to the segments [ɾ] and [l]: there are minimal pairs involving the two sounds ([ɾæm] vs [læm], for instance). In this variety of Scottish English, [ɾ] and [l] are in parallel distribution and can function to signal a semantic contrast. It is important to bear in mind that, when we say that a phonetic difference is contrastive, we refer to a semantic contrast, and not to a phonetic difference between the sounds.

在韩语中，[ɾ] 和 [l] 之间的区别永远不会是对比的，因为 [ɾ] 和 [l] 可能永远不会出现在同一个地方。它们呈互补分布：其中一个出现，另一个从不出现，反之亦然。具体来说，韩语中的 [ɾ] 出现在元音之间但没有出现在其他地方，而 [l]从未出现在元音之间，但可能出现在其他地方。因此，不可能在韩语中找到包含这两个声音的最小对。这两种声音在语音上也很相似：都是浊音，并且都需要在舌叶中心和牙槽嵴之间形成闭合。因此，这两个声音是韩语中同一音素的实现。

In Korean, the distinction between [ɾ] and [l] can never be contrastive, since [ɾ] and [l] may never occur in the same place. They are in complementary distribution: where one occurs, the other never does, and vice versa. Specifically, [ɾ] in Korean occurs between vowels but nowhere else, whereas [l] never occurs between vowels, but may occur elsewhere. Because of this, it is impossible to find minimal pairs involving these two sounds in Korean. The two sounds are also phonetically similar: both are voiced and both entail a closure made between the centre of the tongue blade and the alveolar ridge. Therefore the two sounds are realizations of the same phoneme in Korean.

在这种苏格兰英语中，有 /ɾ/ 与 /l/ 的音位对比。另一方面，在韩语中，没有这样的音位对比：苏格兰英语的这种变体有 /ɾ/ 和 /l/，而韩语只有一个音素：/l/，它有两个音位同位素，[ɾ] 和 [升]。换句话说，[ɾ] 和 [l] 之间的声音差异在苏格兰英语中是音位的，而 [ɾ] 和 [l] 之间的差异是韩语中的同位异音。说这种英语的人将 [ɾ] 和 [l] 视为两个不同的心理类别，而说韩语的人则将它们视为一个单一的心理类别。在韩语中，音位 /l/ 在元音之间实现为 [ɾ]，在其他地方实现为 [l]。

In this variety of Scottish English, there is a phonemic /ɾ/ vs /l/ contrast. In Korean, on the other hand, there is no such phonemic contrast: whereas this variety of Scottish English has /ɾ/ vs /l/, Korean has only one phoneme: /l/, which has two allophones, [ɾ] and [l]. Put another way, the difference between the sounds [ɾ] and [l] is phonemic in Scottish English, whereas the difference between [ɾ] and [l] is allophonic in Korean. Speakers of this variety of English perceive [ɾ] and [l] in terms of two distinct mental categories, whereas Korean speakers perceive them in terms of a single mental category. In Korean, the phoneme /l/ is realized as [ɾ] between vowels, and is realized as [l] elsewhere.

我们可以将这种苏格兰英语和韩语之间的语音差异描述如下：

We may depict this phonological difference between this variety of Scottish English and Korean as follows:

(7) [ɾ]和[l]在苏格兰英语和韩语中的音位地位

我们现在已经说明了说韩语的人的困难所在：在他们（大部分）对他们语言的无意识知识的水平上。就这些片段而言，韩语和苏格兰英语的这种变体在变位音层面上没有区别：都有 [ɾ] 和 [l]。但他们确实在音素层面有所不同：讲苏格兰英语的人有一种讲韩语的人所没有的心理区别；因此，说韩语的人的问题本质上是心理上的（具体来说是知觉上的），而不是发音上的。

We have now shown where the Korean speakers’ difficulty resides: at the level of their (largely) unconscious knowledge of their language. As far as these segments are concerned, Korean and this variety of Scottish English do not differ at the allophonic level: both have [ɾ] and [l]. But they do differ at the phonemic level: the Scottish English speaker has a mental distinction which the Korean speaker lacks; the Korean speakers’ problem is thus mental (specifically, perceptual) in nature, not articulatory.

我们已经说过，在给定的语境中韩语 /l/ 音素的变位音素会出现是完全可以预测的。可以说，以韩语为母语的人无意识地掌握了音位变体的出现，形成了他们语言知识的一部分。我们可以根据语音规则来表达这种概括，如下所示：

We have said that it is entirely predictable which allophone of the Korean /l/ phoneme will occur in a given context. We may say that there is a phonological generalization governing the occurrence of the allophones, which the native speakers of Korean have unconsciously grasped, and which forms part of their linguistic knowledge. We may express that generalization in terms of a phonological rule, as follows:

(8)/l/的实现在韩语中

/l/在元音之间实现为[ɾ]。

正如我们将看到的，母语人士的语言知识包含许多这样的概括。就 [ɾ] 和 [l] 而言，说韩语的人和说苏格兰英语的人的语音知识在两个方面有所不同：(a) 说苏格兰英语的人具有韩语人所没有的语音区别，并且(b) 说韩语的人具有说苏格兰英语的人所缺乏的语音概括。因此，语音知识包括语音类别和语音概括等。

As we will see, the linguistic knowledge of native speakers contains many such generalizations. As far as [ɾ] and [l] are concerned, the phonological knowledge of the Korean speaker and that of the Scottish English speaker differ in two respects: (a) the Scottish English speaker has a phonological distinction which the Korean speaker lacks, and (b) the Korean speaker possesses a phonological generalization which the Scottish English speaker lacks. Phonological knowledge therefore consists of, among other things, phonological categories and phonological generalizations.

在多种英语中，/l/ 音素也有同位异音：'clear l' ([l]) 和 'dark l' ([ɫ])。³以下数据显示了这两种声音在这些品种中的典型分布：

In several varieties of English, the /l/ phoneme also has allophones: ‘clear l’ ([l]) and ‘dark l’ ([ɫ]).3 The following data show the typical distribution of these two sounds in those varieties:

(9) 英文'clear l'和'dark l'

（一种）	[kʰlεvə]	聪明的
（乙）	[bεɫz]	钟声
（C）	[pʰleɪn]	清楚的
(四)	[tɹeɪɪ]	踪迹
(五)	[lʊk]	看
（F）	[pʰʊɫ]	拉
（G）	[lɔ:]	法律
（H）	[bɔ:ɫz]	球
（一世）	[laɪ]	说谎
(十)	[pʰaɪɪ]	桩

说明音位变体分布的一种方法是说“clear l”紧接在元音之前，而“dark l”紧接在元音之后。/l/ 音位与其明、暗变位音位的关系可以用以下规则来表述（稍后我们将用音节结构来表示）：

One way of stating the distribution of the allophones is to say that ‘clear l’ occurs immediately before vowels, whereas ‘dark l’ occurs immediately after vowels. We may state the relationship between the /l/ phoneme and its clear and dark allophones in terms of the following rule (which we will later express in terms of syllable structure):

(10) /l/在英语中的实现

/l/在元音后立即实现为[ɫ]。

我们可以将韩语 /l/ 和 /l/ 在某些英语变体中的实现描述如下：

We may depict the realizations of Korean /l/ and /l/ in certain varieties of English as follows:

(11) /l/ 韩语和英语的实现

5.3 音位、音位变体和上下文

5.3 Phonemes, Allophones and Contexts

我们已经说过，音素的变位音是该音素的可预测实现。给定特定上下文，我们可以预测会出现哪个变位音。在某些情况下，我们引用的上下文类型相当笼统。例如，在我们考虑的韩语数据中，我们看到送气和不送气清塞音都可能出现在单词的开头。我们还看到，在我们查看的韩语数据中，韩语 /l/ 在元音之间实现为 [ɾ]。“在单词的开头”和“元音之间”是非常普遍的上下文。“在单词的末尾”、“辅音之前”或“元音之后”也是如此。

We have said that the allophones of a phoneme are predictable realizations of that phoneme. We can predict which allophone will occur, given a specific context. The sorts of context we have cited are, in some cases, rather general. For instance, in the Korean data we considered, we saw that aspirated and unaspirated voiceless stops may both occur at the beginning of a word. We also saw, in the Korean data that we looked at, that Korean /l/ is realized as [ɾ] between vowels. ‘At the beginning of a word’ and ‘between vowels’ are quite general contexts. So is ‘at the end of a word’, or ‘before a consonant’, or ‘after a vowel’.

在其他情况下，我们需要参考的上下文更为具体。例如，在我们查看的英语数据中，我们看到不送气的清塞音发生在清音牙槽擦音之后。在许多情况下，同位素出现的上下文与同位素本身的性质之间似乎存在某种语音联系。让我们考虑一个例子。

In other cases, the contexts we need to refer to are more specific. For instance, in the English data we looked at, we saw that the unaspirated voiceless stops occurred after a voiceless alveolar fricative. In many cases, there appears to be some kind of phonetic connection between the context in which an allophone occurs and the nature of the allophone itself. Let us consider an example.

在许多英语口音中，/ɹ/ 音素有两种实现方式：[ɹ] 和 [ɹ̥]（其中下标变音符号表示清音）。以下数据说明了这一点：⁴

In many accents of English, the /ɹ/ phoneme has two realizations: [ɹ] and [ɹ̥] (in which the subscript diacritic denotes voicelessness). The following data exemplify this:4

(12) 英语中 /ɹ/ 的浊音和清音变位音

（一种）	[tʰɹ̥aɪ]	尝试
（乙）	[əɹeɪ]	大批
（C）	[pʰɹ̥u:v]	证明
(四)	[gɹoʊ]	生长
(五)	[kʰɹ̥eɪv]	渴望
（F）	[bɹeɪk]	休息
（G）	[fɹ̥i:]	自由
（H）	[dɹɪŋk]	喝
（一世）	[θɹ̥i:]	三
(十)	[bæɹoʊ]	手推车

很明显，浊音和清音的肺泡近似值呈互补分布：清音只出现在清辅音之后，浊音出现在其他地方。问题是我们是否应该说有一个浊齿槽近似音位在清辅音之后被实现为清音位变位，或者说有一个清音齿槽近似音位被实现为浊辅音之后和元音之间的浊音近似值。我们选择前一种说法，因为它在语音上更自然：近似词通常是浊音。此外，我们可以从语音上理解浊音素在清音辅音之后具有清音实现的说法：该实现正在与前面的片段同化（它变得更像是相邻的片段）。

It is clear that the voiced and voiceless alveolar approximants are in complementary distribution: the voiceless one appears only after voiceless consonants, and the voiced one appears elsewhere. The question is whether we should say that there is a voiced alveolar approximant phoneme which is realized as a voiceless allophone after voiceless consonants, or that there is a voiceless alveolar approximant phoneme which is realized as a voiced approximant after voiced consonants and between vowels. We choose the former claim, since it is more phonetically natural: approximants are normally voiced. Additionally, we can make phonetic sense of the claim that a voiced phoneme has a voiceless realization when it follows voiceless consonants: the realization is assimilating to the preceding segment (it is becoming more like an adjacent segment).

让我们考虑另一种情况。在许多英语口音中，塞音位于软腭前，靠近硬腭。以下数据举例说明了这一点（[c] 和 [ɟ] 分别代表清音和浊音腭塞音）：

Let us consider another case of this sort. In many accents of English, there are stops which are articulated in front of the velar place of articulation, close to the hard palate. The following data exemplify this ([c] and [ɟ] represent a voiceless and a voiced palatal stop, respectively):

(13) 英语中的软腭塞音

（一种）	[kʰu:ɫ]	凉爽的
（乙）	[cʰi:p]	保持
（C）	[kʰoʊɫ]	煤炭
(四)	[cʰi:n]	敏锐的
(五)	[kʰɒp]	警察
（F）	[cʰɪt]	成套工具
（G）	[kʰɒ:t]	大车
（H）	[scɪp]	跳过
（一世）	[gu:ɫ]	食尸鬼
(十)	[ɟɪə]	齿轮
(k)	[goʊɫ]	目标
(l)	[ɟiɫ]	鳃

再一次，这两种音段类型呈互补分布：高级腭音出现在高前元音之前，而软腭发音出现在其他地方。我们假设 /k/ 音素在其发音位置“基本上”是软腭音，但在高前元音之前具有前置或高级实现。这在语音上是有道理的：前高元音是硬腭发音（发音器是舌前部和硬腭），所以我们可以说当软腭音素是前高元音时正在同化后面的元音。

Once again, the two segment types are in complementary distribution: the advanced, palatal articulations occur before high front vowels, and the velar ones occur elsewhere. We postulate a /k/ phoneme which is ‘fundamentally’ velar in its place of articulation, but which has a fronted or advanced realization before high front vowels. This makes phonetic sense: high front vowels are palatal articulations (the articulators are the front of the tongue and the hard palate), so we can say that the velar phoneme is assimilating to the following vowel when it is a high front vowel.

我们采纳这样的观点，即音素通常具有一种“默认”或“基本”语音实现，并且这种实现将在没有可将实现从其默认实现“转移”的可指定上下文的情况下发生。

We are adopting the view that phonemes often have a kind of ‘default’ or ‘basic’ phonetic realization, and that it is this realization which will occur in the absence of specifiable contexts which ‘shift’ the realization from its default one.

5.4 总结

5.4 Summing Up

在本章中，我们开始区分语音学，语音学被定义为对语音本身的研究和音韵学，研究与这些声音相关的心理表征、类别和概括的系统。以一种语言为母语的人倾向于认为其语音系统是理所当然的。例如，说英语的人认为 [ɹ] 和 [l] 非常明显，尽管它们在语音上非常相似。同样，说英语的人不能轻易看出 [p] 和 [pʰ] 是不同的，尽管事实上它们是不同的。本章试图表明，这些感知的基础是母语的语音系统，它与以该语言为母语的人发出的语音集不同，如果密切相关的话。听起来是什么认为“相同”或“不同”在很大程度上取决于构成一个人的母语音系的心理范畴系统。但很明显，语音学和音韵学是密切相关的。

In this chapter, we have begun to distinguish between phonetics, defined as the study of speech sounds per se, and phonology, the study of the system of mental representation, categories and generalizations to which those sounds are related. Native speakers of a language tend to take its phonological system for granted. Speakers of English, for instance, think it perfectly obvious that [ɹ] and [l] are quite distinct, despite the fact that they are, phonetically, very similar. Equally, speakers of English cannot easily see that [p] and [pʰ] are different, despite the fact that they are. This chapter has sought to show that what underlies these perceptions is the phonological system of the native language, as distinct from, if intimately related to, the set of speech sounds uttered by native speakers of the language. What sounds one takes to be ‘the same’ or ‘different’ depend to a large extent on the system of mental categories which constitutes one’s native language phonology. But it is clear that phonetics and phonology are intimately connected.

20 世纪初研究北美印第安语言的语言学家爱德华·萨皮尔 (Edward Sapir) 很好地表达了我们心理存储的特定语言语音类别系统在多大程度上支配着我们对语音流的感知：

The extent to which our mentally stored system of language-specific phonological categories governs our perception of a stream of speech sounds was well expressed by the linguist Edward Sapir, who worked with North American Indian languages in the early twentieth century:

未受过教育的语言记录者，只要他有敏锐的听力和真正的语言本能，与容易被大量观察所淹没的精细语音学家相比，往往具有很大的优势。我已经利用我教印度人写他们自己的语言的经验，在另一个方面测试它的价值。它在这里产生了同样有价值的证据。我发现很难或不可能教一个印度人做出与“他的语言模式中的点”不对应的语音区分，但是这些差异可能会打动我们的客观耳朵，但是那些微妙的、几乎听不见的语音差异，如果他们只击中了“模式中的要点”，就可以很容易地、自愿地以书面形式表达出来。在观看我的 Nootka 口译员书写他的语言时，^5个

the unschooled recorder of language, provided he has a good ear and a genuine instinct for language, is often at a great advantage as compared with the minute phonetician, who is apt to be swamped by his mass of observations. I have already employed my experience in teaching Indians to write their own language for its testing value in another connection. It yields equally valuable evidence here. I found that it was difficult or impossible to teach an Indian to make phonetic distinctions that did not correspond to ‘points in the pattern of his language’, however these differences might strike our objective ear, but that subtle, barely audible, phonetic differences, if only they hit the ‘points in the pattern’, were easily and voluntarily expressed in writing. In watching my Nootka interpreter write his language, I often had the curious feeling that he was transcribing an ideal flow of phonetic elements which he heard, inadequately from a purely objective standpoint, as the intention of the actual rumble of speech.5

当一个人考虑到任何一个正常的 6 个月大的孩子，无论他或她开始学习什么语言，都可以区分送气和不送气的清塞音时，人们就可以开始理解一个人的母语音素类别对一个人的感知的影响程度。很明显，送气/不送气的区别原则上可以作为音位区分的基础，而且它在许多人类语言中确实以这种方式起作用。但是，如果孩子学会了一种语言（例如大多数英语变体），其中送气/不送气的区别是同位音而不是音素，那么在一定的意识水平上就会忽略这种区别。因此，获得一个人母语的音系可能会导致一种感知辨别能力的丧失，但仅限于一种意识水平：当说南非英语的人说出不送气塞音而不是送气塞音时，说 RP 的人通常会注意到这一点，即使说 RP 的人只是注意到南方人的讲话有些不同讲非洲英语的人。事实上，这种差异对于使用不送气塞音从不出现在重读元音之前的单词开头的语言的人来说可能是相当惊人的。这样的说话者，在突然遇到带有希腊口音的英语时（例如，在到达希腊机场时）通常会认为像这样的词即使说 RP 的人只是注意到说南非英语的人的讲话有些不同。事实上，这种差异对于使用不送气塞音从不出现在重读元音之前的单词开头的语言的人来说可能是相当惊人的。这样的说话者，在突然遇到带有希腊口音的英语时（例如，在到达希腊机场时）通常会认为像这样的词即使说 RP 的人只是注意到说南非英语的人的讲话有些不同。事实上，这种差异对于使用不送气塞音从不出现在重读元音之前的单词开头的语言的人来说可能是相当惊人的。这样的说话者，在突然遇到带有希腊口音的英语时（例如，在到达希腊机场时）通常会认为像这样的词盖特威克机场（伦敦机场之一）的发音为 [gadwig]。在这种情况下，说英语的人不仅意识到所讨论的塞音是不送气的，而且还将它们归类为英语浊塞音，因为英语中的浊塞音是不送气的，并且词首和词尾英语中的浊音塞子几乎没有浊音。

One can begin to appreciate the extent to which one’s native language phonemic categories affect one’s perception when one considers that any normal 6-month-old child, no matter what language he or she is beginning to acquire, can distinguish aspirated and unaspirated voiceless stops. Clearly, then, the aspirated/unaspirated difference is one that could in principle act as the basis for a phonemic distinction, and it clearly does act that way in many human languages. But a child who acquires a language (such as most varieties of English) in which the aspirated/unaspirated distinction is allophonic rather than phonemic will come to ignore that distinction at a certain level of awareness. Acquiring the phonology of one’s native language can therefore result in a kind of loss of perceptual discrimination, but only at one level of awareness: when a speaker of, say, South African English utters unaspirated stops instead of aspirated stops, this will often be noticed by a speaker of, say, RP, even if the RP speaker notices only that there is something different about the speech of the South African English speaker. Indeed, such differences can be quite striking to the speaker of a language in which unaspirated stops never occur word-initially before a stressed vowel. Such speakers, on being suddenly confronted by English spoken with, say, a Greek accent (on arrival, say, at a Greek airport) will typically think that a word such as Gatwick (one of the London airports) is being pronounced as [gadwig]. In cases such as this, the English speaker not only perceives the fact that the stops in question are unaspirated, but also assigns them to the category of English voiced stops, because voiced stops in English are unaspirated, and word-initial and word-final voiced stops in English are barely voiced at all.

通过比较和对比，以英语为母语的人和成年英语学习者都可以开始意识到她或他自己的语音系统，以及这对一个人对语音的感知的巨大影响就某些声音集而言，语音上相同（或几乎相同）但语音上不同的语言。本章中给出的示例旨在开始引发这种意识，随后的练习也是如此。

Both the native speaker and the adult learner of English can begin to develop an awareness of her or his own phonological system, and of the immense influence this has on one’s perception of speech sounds, by comparing and contrasting languages which are phonetically identical (or nearly identical), but phonologically distinct, with respect to some set of sounds. The examples given in this chapter are designed to begin to induce this kind of awareness, as are the exercises which follow.

笔记

Notes

1这些数据并未显示大多数英语口音中出现送气和不送气清塞音的全部位置。因此，我们不得不说的关于它们的音位地位的内容过于简单化了。但数据足以说明一个有效的观点。

1 These data do not show the full range of places in which aspirated and unaspirated voiceless stops occur in most English accents. What we will have to say about their phonological status is therefore very much oversimplified. But the data will suffice to illustrate a valid point.

2韩语有第三类塞音，我们将在下面讨论。

2 Korean has a third phonemic category of stops, which we discuss below.

3 /l/ 也有清音变位音；我们在这里忽略这些。

3 There are also devoiced allophones of /l/; we ignore these here.

4我们在这里表示这些清音，但从今以后，如果清音现象与所提出的观点无关，我们将不会使用“清音”变音符号来转录它们。

4 We indicate these devoiced sounds here, but henceforth we will not transcribe them using the ‘voiceless’ diacritic in cases where the devoicing phenomenon is irrelevant to the point being made.

5 Edward Sapir (1921)，《语言》，纽约：Harcourt Brace，p. 56.

5 Edward Sapir (1921), Language, New York: Harcourt Brace, p. 56.

练习

Exercises

1 英语和西班牙语的 [d] 和 [ð]

(一) 英语

1.	[dεn]	书房
2.	[ðεn]	然后
3.	[doʊz]	瞌睡
4.	[ðoʊz]	那些
5.	[dεə]	敢
6.	[ðεə]	他们的
7.	[ʌdə]	乳房
8.	[ʌðə]	其他
9.	[aɪdə]	羽绒
10.	[aɪðə]	任何一个

正如这些数据所示，在许多（并非全部）英语口音中，[d] 和 [ð] 是不同音素的实现。它们是平行分布的（都出现在单词的开头和元音之间）。它们的功能也相反：涉及这两种声音的配对最少。因此，我们有理由为大多数英语口音假设 /d/ 与 /ð/ 的音位区别。

(二) 西班牙语

现在考虑以下西班牙数据。（有问题的浊音停顿实际上是西班牙语中的牙音。我们在这里忽略了这一事实。）[d] 和 [ð] 之间的区别是西班牙语中的音位还是同位异音？用证据和论证证明你的答案。

1.	[dɐr]	'给'
2.	[nɐðɐ]	'没有什么'
3.	[dεβεr]	'必须'
4.	[boðeɣɐ]	'酒窖'
5.	[diɐs]	'天'
6.	[ɐblɐðo]	'说'
7.	[bɐndɐ]	'丝带'
8.	[pɾɐðo]	'草地'
9.	[ɐndɐr]	'去'
10.	[poðεr]	'能够'

2 英语和韩语语音站

在韩语中，/p/、/t/ 和 /k/ 在词尾（如以下数据所示）或另一个辅音之前有未释放的变位音。/p/、/t/ 和 /k/ 音素也有浊塞音同位异音：[b]、[d] 和 [g]。与英语不同，韩语没有浊音塞音：[b]、[d] 和 [g] 在韩语中始终是 /p/、/t/ 和 /k/ 的同位异音。检查以下数据并说出浊塞音变位音出现的上下文：

gueq05_076

英语和标准阿拉伯语的 3 个声门塞

语音片段 [ʔ]（声门塞音）出现在大多数说英语的人的语音中，但英语中没有声门塞音（/ ʔ/），因为 [ʔ] 从来没有与任何其他片段形成对比。例如，[kʰεʔəə]和[kʰεtəə]（kettle）不是不同词的读音，而是同一个词的不同读法。以下是标准阿拉伯语的一些数据。这种语言有声门塞音吗？解释你的答案背后的原因：

（一种）	[faʔl]	'福'
（乙）	[致命]	'扭曲/缠绕'
（C）	[faʔɾ]	'老鼠'
(四)	[fa:ɾ]	“煮沸了”
(五)	[baʔs]	'力量'
（F）	[ba:s]	“他吻了”
（G）	[buʔs]	'苦难'
（H）	[公共汽车]	'公共汽车'

gueqrecard_001

4 进一步的音标练习

在www.wiley.com/go/carrphonetics收听曲目 5.1。尽可能详细地转录您在录音中听到的单词，注意细节，例如存在与不存在送气、清晰与黑暗 l 以及 /l/ 和 /ɹ/ 的清音同位异音。

开始玩价格可能会杀死 clear creep

6个

6 英语音素

English Phonemes

6.1 英语辅音音素

6.1 English Consonant Phonemes

我们已经将音素与语音段区分开来，并开始提出关于哪些音素可能作为母语人士的语音知识的一部分存在的假设。具体来说，我们已经说过，许多说英语的人都有辅音音素 /l/、/ɹ/、/p/、/t/ 和 /k/ 等。我们将很快假设说英语的人拥有一个完整的辅音音素系统。但我们必须首先更准确地理解“说英语的人”是什么意思。显然，英语有不同的变体，我们稍后会更详细地考虑，我们需要一些区分它们的方法。让我们首先考虑语言学家经常呼吁的一个区别：口音和方言之间的区别. 人们常说，口音的差异仅涉及语音和音系的变化，而方言的差异不仅如此：它们还包括词汇和句法的差异。这是一种相当简单的区分方式，也是一种充满困难的区分方式，但对于目前的讨论来说已经足够了。

We have distinguished phonemes from phonetic segments, and have begun to formulate hypotheses about which phonemes might exist as part of the native speaker’s phonological knowledge. Specifically, we have said that many English speakers have the consonant phonemes /l/, /ɹ/, /p/, /t/ and /k/, among others. We will shortly postulate a full system of consonant phonemes which English speakers have. But we must first be a little more precise about what we mean by ‘English speakers’. Clearly, there are different varieties of English, which we will be considering in more detail later, and we will need some means of differentiating between them. Let us begin by considering a distinction which is often appealed to by linguists: that between accent and dialect. It is often said that differences in accent concern solely phonetic and phonological variation, whereas dialect differences involve more than this: they also include differences in vocabulary and syntax. This is a rather simplistic way of putting the distinction, and it is a distinction which is fraught with difficulties, but it will suffice for the present discussion.

我们可以举例说明口音和方言之间的区别如下。也许最广泛使用（和书写）的英语方言是被称为标准英语的“声望”方言，它起源于英格兰东南部；这种方言在英国、国家广播和电视台、新闻界以及大多数印刷出版物中被广泛使用。可以说带有新西兰口音、泰恩赛德口音、纽约市口音或任何英语口音的标准英语。当发生这种情况时，我们可以说（稍微简化）说话者正在使用标准英语的词汇和句法，同时保留构成母语口音的语音和音系。

We may exemplify the difference between accent and dialect as follows. Perhaps the most widely spoken (and written) English dialect is the ‘prestige’ dialect known as Standard English, which has its origins in the South East of England; this dialect is used widely, in Britain, in national radio and television, in the press, and indeed in most printed publications. It is possible to speak Standard English with a New Zealand accent, a Tyneside accent, a New York City accent, or indeed any accent of English. When this happens, we may say (simplifying somewhat) that the speaker is using the vocabulary and syntax of Standard English, while retaining the phonetics and phonology which constitutes the native accent.

让我们更详细地举例说明方言和口音之间的区别，如下所示。拿标准的英语句子You will not be able to put the children on the floor. 由具有标准苏格兰英语 (SSE) 口音的说话者说出，结果将是：

Let us exemplify the difference between dialect and accent in a little more detail, as follows. Take the Standard English sentence You will not be able to put the children on the floor. Uttered by a speaker with a Standard Scottish English (SSE) accent, the outcome would be:

(1) [jəɫnɒʔbiebɫtəpʰʉʔðətʃɪɫdɹənənðəflo:ɹ]

现在将此与 RP 的说话者说出的相同标准英语句子进行比较。RP 发言人可能会说：

Now compare this with the same Standard English sentence uttered by a speaker of RP. The RP speaker might well utter:

(2) [jəənɒʔbieɪbɪtəpʰʊʔðətʃɪɫdɹənənðəflɔ:]

两位说话者都说标准英语（句法相同，词汇也一样，如果排除词汇中语素的音韵形式），但他们的口音不同：SSE 说话者的元音并不总是与英语相同RP 扬声器，而 SSE 扬声器在floor中发出 [ɹ] ，而 RP 扬声器则不会。

Both speakers are speaking Standard English (the syntax is the same, as is the vocabulary, if one excludes the phonological form of the morphemes in that vocabulary), but their accents differ: the SSE speaker’s vowel sounds are not always identical to those of the RP speaker, and the SSE speaker utters an [ɹ] in floor, which the RP speaker does not.

现在让我们想象一下，同一位 SSE 演讲者想要传达相同的主张，但这次他使用的是标准英语以外的方言：低地苏格兰语。结果可能是：

Now let us imagine that the same SSE speaker wants to convey the same proposition, but speaking, this time, in a dialect other than Standard English: that of Lowland Scots. The result might be:

(3) [jʌɫno:kɪmpʰɪʔðəbe:ɹɪnzonðəfɫe:ʌɹ]

（这可能写成你不能把婴儿放在地板上。）

(This might be written as You’ll no can put the bairns on the floor.)

在（3）中，句法和词汇与标准英语不同；我们可以说这些是方言差异，并且与我们在上面 (1) 和 (2) 之间注意到的口音差异在种类上截然不同。

In (3), the syntax and vocabulary differ from that of Standard English; we may say that these are dialectal differences, and distinct in kind from the differences in accent which we noted between (1) and (2) above.

我们将在后面的章节中回到口音变化的问题；现在，让我们看看大多数英语变体共有的辅音音位系统，通常如下所示：

We will return to the matter of accent variation in a later chapter; for the moment, let us look at the consonant phoneme system shared by most varieties of English, which typically looks like this:

(4) 英语辅音音素

/p/	如pie, pit, rip
/乙/	如buy, bit, rib
/吨/	如领带、小费、令状
/d/	如死、浸、除
/k/	如cool, kit, rick
/G/	就像食尸鬼，git，钻机
[tʃ]	就像咀嚼，chit，rich
[dʒ]	如犹太人，杜松子酒，山脊
/θ/	如大腿，瘦，有
/ð/	就像那时一样，那个，镰刀
/F/	如fie, fit, riff
/v/	就像维恩，增值税，离开
/秒/	如叹息、坐下、租赁
/z/	就像在动物园一样，请压缩
/H/	就像在高处，臀部
/ʃ/	如害羞、船舶、皮带、网格
/ʒ/	作为测量
/w/	就像在湿地一样，赢了
/升/	就像谎言、嘴唇、真实
/ɹ/	就像黑麦一样，撕裂
/ε/	和去年一样
/米/	就像在我的，肉，边缘
/n/	如nig, neat, sin
/ŋ/	如唱歌，响铃

证据部分以 (4) 中引用的最小对的形式出现，例如measure/ mesher ([mεʒə]/[mεʃə]) 和 sigh /shy ([saɪ]/[ʃaɪ])，但我们有决不会在这里提供所有证据。让我们简要地看一下这里假设的三个鼻音塞音的一些证据：

The evidence comes partly in the form of the sorts of minimal pair cited in (4), such as measure/mesher ([mεʒə]/[mεʃə]) and sigh/shy ([saɪ]/[ʃaɪ]), but we have by no means presented all of the evidence here. Let us look briefly at some of the evidence for the three nasal stop phonemes postulated here:

(5) 英语鼻音塞音的证据

（一种）	[mi:t]	肉
（乙）	[ni:t]	整洁的
（C）	[moʊɫ]	痣
(四)	[否ʊɫ]	小丘
(五)	[sɪn]	罪
（F）	[sɪŋ]	唱歌
（G）	[dɪm]	暗淡
（H）	[dɪn]	喧嚣
（一世）	[wɪn]	赢
(十)	[wɪŋ]	翅膀

很明显 [m] 和 [n] 是平行分布的：每个都可能出现在词首或词末。这种区别也是对比的：它构成了最小对的基础，例如meat/neat。还可以看出，[n]和[ŋ]是平行分布的：虽然[ŋ]没有出现在词首，但都可能出现在词尾。这种区别也是对比性的，如win/wing等最小对的存在所示。[m] 和 [ŋ] 之间的区别也是对比性的，例如whim/wing展示。因此，我们有明确的证据表明 /m/、/n/ 和 /ŋ/ 之间存在三向音位区别。我们将在下面更详细地考虑这种分析。目前要说明的重点是，我们在证据和论证的基础上假设音素的存在；如果音素是心理范畴，则无法直接观察到它们。

It is clear that [m] and [n] are in parallel distribution: each may occur word-initially or word-finally. The distinction is also contrastive: it forms the basis for minimal pairs such as meat/neat. It is also clear that [n] and [ŋ] are in parallel distribution: while [ŋ] does not appear in word-initial position, both may occur in word-final position. The distinction is also contrastive, as is shown by the existence of minimal pairs such as win/wing. The distinction between [m] and [ŋ] is contrastive too, as pairs such as whim/wing show. We therefore have clear evidence for a three-way phonemic distinction between /m/, /n/ and /ŋ/. We will consider this analysis in more detail below. The main point to be made at the moment is that we postulate the existence of phonemes on the basis of evidence and argumentation; if phonemes are mental categories, they cannot be directly observed.

6.2 语素的音系形式

6.2 The Phonological Form of Morphemes

我们已经说过，在了解一种语言时，说话者在很大程度上拥有无意识的语言知识，这些知识包括语义、句法和语音知识。我们已经说过，我们称为音素的音位单位或类别是音位知识的一部分。随着我们在本书中的进步，我们将研究说话者除了音素之外还拥有哪些其他类型的语音知识的问题。让我们从单词的内部结构开始这项研究。你会同意英文单词cats可以分解成两个组成部分。让我们称这些组成部分为语素。那么我们可以说这个词由一个词根词素和一个复数词素组成（在这种情况下，它是一个后缀）。让我们说这类词在形态上很复杂: 它们由不止一个语素组成。让我们假设语素采用三元组的形式：句法、语义和音位。以语素cat为例：它有一个句法（它是一个名词）、一个语义（它的意思是“猫”）和一个音系，其形式为 /kæt/；我们将把它称为语素的音系形式。显然，语素的音位形式可能由不止一个音素组成。正如音素是心理对象一样，这个语素的音系形式也是心理对象：/kæt/ 是说话者头脑中的心理表征，而序列 [kʰæt] 是语音序列。

We have said that, in knowing a language, a speaker possesses largely unconscious linguistic knowledge, which subsumes semantic, syntactic and phonological knowledge. And we have said that the phonological units or categories we have called phonemes are part of that phonological knowledge. As we progress in this book, we will investigate the question of what other sorts of phonological knowledge speakers possess, besides phonemes alone. Let us begin this investigation by considering the internal structure of words. You will agree that the English word cats may be broken down into two component parts. Let us call those component parts morphemes. Then we may say that this word consists of a root morpheme and a plural morpheme (which, in this case, is a suffix). Let us say that words of this sort are morphologically complex: they consist of more than one morpheme. Let us say that a morpheme takes the form of a triple: a syntax, a semantics and a phonology. Take the morpheme cat: it has a syntax (it is a noun), a semantics (it means ‘cat’) and a phonology, which takes the form /kæt/; we will refer to this as the phonological form of the morpheme. The phonological form of a morpheme may, clearly, consist of more than one phoneme. Just as phonemes are mental objects, so the phonological form of this morpheme is a mental object: /kæt/ is a mental representation in the mind of a speaker, whereas the sequence [kʰæt] is a phonetic sequence.

现在让我们考虑形容词不可能的、不平衡的、不恰当的、无形的、间接的、疯狂的、不正确的和不光彩的。所有词都至少包含一个前缀词素和一个词根词素（其中一些词有一个前缀、一个词根和一个后缀）。许多说话者对这些词有以下发音：

Let us now consider the adjectives impossible, imbalanced, infelicitous, intangible, indirect, insane, incorrect and inglorious. All consist of at least a prefix morpheme and a root morpheme (some of these words have a prefix, a root and a suffix). Many speakers have the following pronunciations of these words:

(6)

（一种）	[ɪmpʰɒsɪbɪ]	不可能的
（乙）	[ɪmbælənst]	不平衡的
（C）	[iɱfəlɪsɪtəs]	不合时宜的
(四)	[ɪntʰændʒɪbɪ]	无形的
(五)	[ɪndɪɹεkt]	间接
（F）	[ɪnseɪn]	疯狂的
（G）	[ɪŋkəɹεkt]	不正确的
（H）	[ɪŋglɔ:ɹiəs]	不光彩的

这些词都具有相同的前缀，这是母语人士无意识的英语语言知识的一部分。该前缀是英语的语素之一，并且与该语言中的所有语素一样，具有句法（它是一个前缀）、语义（它具有特定的含义）和音系。但是那个语素的音韵形式是什么？我们从资料中得知，后缀可以实现为[ɪm]、[ɪɱ]、[ɪn]或[ɪŋ]。很明显，前缀中的第一个音素是 /ɪ/，第二个音素是鼻音，但是鼻音素？上面我们说英语有三个鼻音素：/m/、/n/和/ŋ/。所以这个前缀的音韵形式可能是 [ɪm]、[ɪn] 或 [ɪŋ]。让我们考虑 [ɪŋ]。我们可以说 /ŋ/ 音素在 /t/、/d/ 和 /s/ 之前实现为 [n]，在 [p] 和 [b] 之前实现为 [m]。这似乎是有道理的：我们可以说，当前缀添加到词根时，鼻音的发音位置与词根中第一个辅音的发音位置相同。因此，当它后面跟着牙槽辅音时是牙槽辅音（例如 /t/、/d/ 和 /s/）、唇齿辅音（例如 /f/）和双唇辅音（例如 /p/ 或/b/). 这就是我们在第 2 章中提到的同化过程，在这个过程中，当两个部分相邻时，一个部分在某些方面变得与另一个部分相似。在这里，同化代替了发音。进一步的证据表明英语中的鼻音经过发音部位同化并不难获得。考虑以下数据，这些数据代表了许多说英语的人的演讲：

It is part of the native speaker’s unconscious linguistic knowledge of English that these words all have the same prefix. That prefix is one of the morphemes of English, and, like all morphemes in the language, has a syntax (it is a prefix), a semantics (it has a specific meaning) and a phonology. But what is the phonological form of that morpheme? We know from the data that the suffix may be realized as [ɪm], [ɪɱ], [ɪn] or [ɪŋ]. It is clear, then, that the first phoneme in the prefix is /ɪ/ and the second phoneme is a nasal, but which nasal phoneme? We claimed above that English has three nasal phonemes: /m/, /n/ and /ŋ/. So the phonological form of this prefix might be [ɪm], [ɪn] or [ɪŋ]. Let us consider [ɪŋ]. We could say that the /ŋ/ phoneme is realized as [n] before /t/, /d/ and /s/, and as [m] before [p] and [b]. This seems to make sense: we can say that, when the prefix is added to a root, the place of articulation of the nasal becomes identical to that of the first consonant in the root. Thus, it is alveolar when followed by an alveolar consonant (such as /t/, /d/ and /s/), labio-dental when followed by a labio-dental consonant (such as /f/), and bilabial when followed by a bilabial consonant (such as /p/ or /b/). This is the process of assimilation we referred to in chapter 2, in which one segment becomes similar, in some respect, to another when the two are adjacent. Here, the assimilation is in place of articulation. Further evidence that nasals in English undergo place of articulation assimilation is not hard to come by. Consider the following data, which are representative of the speech of many speakers of English:

(7) 英语鼻音同化

（一种）	[ʌŋkʰlɪə]	不清楚
（乙）	[ʌŋgɒdli]	不敬虔的
（C）	[ʌɱfεə]	不公平
(四)	[ʌɱvælju:d]	未估价
(五)	[ʌntʰɹu:]	不真实的
（F）	[ʌndʌn]	撤消
（G）	[ʌmbεəɹəbəə]	难以忍受
（H）	[ʌmbaɪəst]	不偏不倚

虽然 [ɪŋ] 的解决方案是合理的，但它面临一个困难：我们同样可以说语素的音系形式是 [ɪn] 或 [ɪm]，并且在任何一种情况下，鼻音都同化为后面的辅音。根据目前提供的证据，在这三种选择之间没有非任意的选择方式：每一种都和其他的一样合理。但是，以下数据允许我们做出非任意选择：

While the [ɪŋ] solution is plausible, it faces a difficulty: we might equally say that the phonological form of the morpheme is [ɪn], or [ɪm], and that, in either case, the nasal assimilates to a following consonant. On the evidence presented thus far, there is no non-arbitrary way of choosing between the three alternatives: each is as plausible as the others. The following data, however, allow us to make a non-arbitrary choice:

(8)

（一种）	[ɪæktɪv]	不活跃
（乙）	[ɪnɐpɹətɪv]	无效
（C）	[ɪnεfəbɪ]	无法形容的
(四)	[ɪnədvaɪzəbɪ]	不宜
(五)	[ɪnɔ:dɪbɪ]	听不见
（F）	[ɪneɪliənəbɪ]	不可剥夺的

在每种情况下，在词根的开头都没有鼻音可以同化的辅音：每个词根都以元音开头。根据元音词根以 [ɪn] 形式实现的事实，我们因此可以得出结论，前缀的音韵采用 [ɪn] 形式，并且鼻音不会改变其发音位置，如果词根词首段是元音或牙槽辅音。请注意，这通常适用于英语中的牙槽鼻音，因为以下数据（涉及（7）中的前缀）表明：

In each case, there is no consonant at the beginning of the root to which the nasal could assimilate: each root begins with a vowel. From the fact that vowel-initial roots are realized with the [ɪn] form, we can therefore conclude that the phonology of the prefix takes the form [ɪn], and that the nasal does not change its place of articulation if the root-initial segment is a vowel or an alveolar consonant. Note that this is generally true of alveolar nasals in English, as the following data, involving the prefix seen in (7), suggest:

(9)

（一种）	[ʌneɪdəd]	独立的
（乙）	[ʌnətʰɹæktɪv]	没有吸引力
（C）	[ʌnivεntfəə]	平安无事
(四)	[ʌnɔ:θədɒks]	非正统的

当然，我们可能会说所讨论的词素有四种不同的音位形式：/ɪm/、/ɪɱ/、/ɪn/ 和 /ɪŋ/，以及impossible等词，infelicitous、indirect和incorrect都在心理上存储了适当的前缀。这种方法有两个问题。首先，没有独立的证据表明英语中有 /ɱ/ 音素（[ɱ] 从来没有与任何其他鼻音形成对比）。其次，即使没有 [ɪɱ] 形式，我们也将致力于根据“多种音位形式”的方法声称/ɪm/ 形式仅附加在后面纯属巧合对于以双唇辅音开头的词根，/ɪŋ/ 仅适用于以软腭辅音开头的词根，而 /ɪn/ 仅适用于以齿龈辅音和元音开头的词根。但这肯定是一个难以置信的说法。因此，至少对于这种情况，我们应该为一个语素假定不止一种音系形式的想法是非常没有吸引力和难以置信的。鉴于我们目前所看到的数据，说任何给定的语素都具有单一的音位形式似乎更为合理。如果是这样的话，那么假设这种形式可能是什么就是音韵学家的任务了。这样做时，她或他将以证据和论证为指导：事情的事实，因为它们本质上是精神的，因此不能直接观察到，所以不能通过五种感官立即检查。

We might, of course, have said that the morpheme in question has four different phonological forms: /ɪm/, /ɪɱ/, /ɪn/ and /ɪŋ/, and that words such as impossible, infelicitous, indirect and incorrect are each stored mentally with the appropriate prefix. There are two problems with this approach. Firstly, there is no independent evidence that there is an /ɱ/ phoneme in English ([ɱ] never functions contrastively with any other nasal). Secondly, even if there were no [ɪɱ] forms, we would be committed to claiming, under the ‘several phonological forms’ approach, that it is mere coincidence that the /ɪm/ form is attached only to roots beginning with a bilabial consonant, the /ɪŋ/ form only to roots beginning with a velar consonant, and the /ɪn/ form only to roots beginning with alveolar consonants and vowels. But that is surely an implausible claim. So, for this sort of case at least, the idea that we should postulate more than one phonological form for a morpheme is deeply unattractive and implausible. Given the data we have seen thus far, it appears much more plausible to say that any given morpheme has a single phonological form. And if that is the case, then it is the phonologist’s task to hypothesize as to what that form might be. In doing so, she or he will be guided by evidence and argumentation: the facts of the matter, since they are mental in nature, and thus not directly observable, will not be available for immediate inspection via the five senses.

在采用这种“每个词素一个音位形式”的方法时，我们允许，虽然任何给定的词素只有一个音位形式，但该音位形式可能在某种意义上“对应”各种不同的语音形式。在我们刚刚看到的例子中，前缀的音位形式为 [ɪn]，但它又对应于四种不同的音位形式：[ɪm]、[ɪɱ]、[ɪn] 和 [ɪŋ]。音韵学家将这种语音形式称为交替：我们说四种形式之间存在交替。给定词中出现给定词素的哪个替代词是完全可以预测的; 有一种概括可以捕捉到这种可预测性，我们能够以语音规则的形式表达它，就像我们在韩语中对 [ɾ] 和 [l] 所做的那样。在我们的英语案例中，所讨论的规则涉及一般的鼻音；它可以非正式地表述为：

In adopting this ‘one phonological form per morpheme’ approach, we are allowing that, while any given morpheme has only one phonological form, that phonological form may ‘correspond’ in some sense to a variety of different phonetic forms. In the case we have just looked at, the prefix has the phonological form [ɪn], but that in turn corresponds to four different phonetic forms: [ɪm], [ɪɱ], [ɪn] and [ɪŋ]. Phonologists refer to such phonetic forms as alternants: we say that there is an alternation between the four forms. Which alternant of a given morpheme occurs in a given word is entirely predictable; there is a generalization which captures that predictability, and we are able to express it in the form of a phonological rule, just as we did with [ɾ] and [l] in Korean. In our English case, the rule in question concerns nasals in general; it might be put informally as:

(10)英语鼻音同化规律

如果前缀的语音形式以鼻音结尾，那么该鼻音将同化到后面辅音的发音位置。

我们可以将这种概括表述为形式化规则，或者作为对英语语素音位形式的某种约束。我们不会深入探讨表达这种概括所需的形式主义类型，也不会探究它们最好用规则还是约束来表达。最重要的一点是，以母语为母语的人似乎拥有这种概括，而且这些概括似乎构成了他们很大程度上无意识的语音知识的一部分。

We could have formulated this generalization as a formalized rule, or as some kind of constraint on the phonological form of morphemes in English. We will not go into the types of formalism required to express such generalizations, or inquire whether they are best expressed as rules or as constraints. The most important point is that native speakers appear to be in possession of generalizations of this sort, and that these appear to constitute a part of their largely unconscious phonological knowledge.

我们刚刚考虑的数据也例证了一个重要的现象：音位重叠。根据（5）中的数据，我们假设了以下鼻塞音音素，实现如下：

The data we have just considered also exemplify an important phenomenon: that of phonemic overlapping. On the basis of the data in (5), we postulated the following nasal stop phonemes, with the realizations shown:

(11)

然而，我们已经允许 /n/ 音素如果在双唇辅音之前也可以实现为 [m]，或者如果它在软唇辅音之前则可以实现为 [ŋ]。这意味着 [m] 的给定出现，例如，可能是 /m/ 的实现，如map中，或 /n/ 的实现，如improbable中。也就是说，/m/ 和 /n/ 音素在它们的实现中是重叠的。我们可以这样描述：

However, we have allowed that the /n/ phoneme may also be realized as [m] if it precedes a bilabial consonant, or [ŋ] if it precedes a velar consonant. This means that a given occurrence of [m], for instance, may be either a realization of /m/, as in map, or a realization of /n/, as in improbable. That is, the /m/ and /n/ phonemes overlap in their realizations. We may depict this as follows:

(12)

问题出现了：说英语的人如何判断给定的 [m] 是 /m/ 还是 /n/ 的实现？答案是语音上下文允许说话者告诉：不在双唇辅音之前的 [m] 将是 /m/ 音素的实现。/m/ 和 /n/ 之间的音位对比据说在双唇辅音之前被中和了。中和是在一个或多个可指定的上下文中暂停音位对比。

The question arises: how can the speaker of English tell whether a given [m] is a realization of /m/ or of /n/? The answer is that the phonological context allows the speaker to tell: an [m] which does not precede a bilabial consonant will be a realization of the /m/ phoneme. The phonemic contrast between /m/ and /n/ is said to be neutralized before a bilabial consonant. Neutralization is the suspension of phonemic contrasts in one or more specifiable contexts.

6.3 英语元音音素

6.3 English Vowel Phonemes

英语口音在其元音音位系统和这些音位所具有的音位变体范围内差异很大。我们首先为 RP 和 GA 描述一组假设的元音音素。我们在第 3 章和第 4 章中介绍和讨论的 RP 和 GA 语音元音质量对于大多数使用这些口音的人来说通常是对比的，因此我们可以为 RP 和 GA 假设以下（重读）元音音素：

Accents of English vary considerably in their vowel phoneme systems and in the range of allophones that those phonemes have. We begin by depicting a set of postulated vowel phonemes for RP and GA. The RP and GA phonetic vowel qualities we presented and discussed in chapters 3 and 4 are typically contrastive for most speakers of those accents, and we may therefore postulate the following (stressed) vowel phonemes for RP and GA:

(13a)

RP元音音素		Wells（1982；参见推荐阅读）词汇集
/ʌ/	就像推杆一样	支撑
/ʊ/	作为投入	脚
/你：/	如游泳池、鞋子	鹅
/ɪ/	就像在坑里	成套工具
/一世：/	如在泥炭中，lea	羊毛
/ε/	作为宠物	裙子
/eɪ/	就像在头上一样，躺在	脸
/ɒ/	像在锅里	很多
/oʊ/	如极点，低	山羊
/ɔ:/	就像在港口一样，法律	北方，力量，思想
/æ/	就像拍拍一样	陷阱
/ɒ:/	作为一部分，Shah	开始，沐浴，手掌
/ɜ:/	如在pert, 毛茸茸的	护士
/ɔɪ/	就像硬币一样，男孩	选择
/aɪ/	作为一堆，买	价格
/aʊ/	噘嘴，牛	嘴
/ɪə/	就像凶猛的，leer	靠近
/εə/	就像稀缺的巢穴	正方形
/ʊə/	就像在葫芦里一样，引诱	治愈

(13b)

GA元音音素	Wells (1982) 词汇集
/ʌ/ 就像推杆一样	支撑
/ʊ/ 如输入	脚，治愈
/u:/ 如泳池、鞋子	鹅
/ɪ/ 就像在坑里一样	成套工具
/i:/ 如在泥炭中，lea	羊毛，近
/ε/ 如宠物	连衣裙，正方形
/eɪ/ 如pate, lay	脸
/oʊ/ 如极点，低	山羊
/ɔ:/法律上的简称	北方，力量，思想
/æ/ 如pat	陷阱，沐浴
/ɒ/ 作为部分，Shah，pot，caught	开始，手掌，很多
/ɜ/ 如pert, furry	护士
/ɔɪ/ 如硬币，男孩	选择
/aɪ/ 如在pile 中，购买	价格
/aʊ/噘嘴，奶牛	嘴

同样，RP 或 GA 元音音素的集合可能是基于证据和一般理论考虑的论证问题。例如，我们可能会建议像pew这样的单词中的第二个元音是双元音 ([iu:]) 并且因为pew与pie、pea等形成最小对，那么 /iu:/ 是一个 RP 元音音素。我们将在适当的时候回到这类问题。我们还应该注意到这里没有列出另一个元音音素：/ə/ (schwa)，它不同于上面列出的所有其他音素，因为它不出现在重读位置（正如我们在第 3 章中提到的） . 我们还将在适当的时候返回到 /ə/ 及其与语音段 [ə] 的关系。

Again, what the set of RP or GA vowel phonemes might be is a matter for argumentation based on evidence and general theoretical considerations. For instance, we might have suggested that the second vowel in words like pew is a diphthong ([iu:]) and that, since pew forms a minimal pair with pie, pea, etc., then /iu:/ is an RP vowel phoneme. We will return, in due course, to this kind of question. We should also note that there is a further vowel phoneme which is not listed here: /ə/ (schwa), which differs from all the other phonemes listed above, since it does not occur in stressed position (as we noted in chapter 3). We will also return, in due course, to /ə/ and its relation to the phonetic segment [ə].

与辅音音素一样，元音音素可能有变位音。例如，说许多英语口音的人对元音音素 /i:/ 有两种实现：[i:] 和 [i:ə]。后者通常发生在软化侧线（'dark l'）之前，如以下数据所示：

Like consonant phonemes, vowel phonemes may have allophones. For instance, speakers of many accents of English have two realizations of the vowel phoneme /i:/: [i:] and [i:ə]. The latter typically occurs before a velarized lateral (‘dark l’), as the following data show:

(14) /i:/ 的同位异音

（一种）	[合身]	脚
（乙）	[fi:əɫ]	感觉
（C）	[蘸]	深的
(四)	[di:əɫ]	交易
(五)	[pʰi:k]	顶峰
（F）	[pʰi:əɫ]	剥
（G）	[si:m]	似乎
（H）	[si:əɫ]	海豹

我们假设 /i:/ 而不是 /i:ə/ 作为音素的形式，因为我们假设当它出现在黑暗 l 之前时，音素的实现受到黑暗 l 的影响。从高前 [i:] 位置缩回的 schwa 发音是元音发音同化到暗 l 中舌体缩回的问题。在这样做时，我们诉诸语音动机的想法：我们的分析是语音动机的，因为我们可以为 /i:/ → [i:ə] 过程提供发音原因，而我们无法提供任何/i:ə/ → [i:] 词尾和除 [ɫ] 之外的任何辅音之前的过程的这种动机。

We postulate /i:/ rather than /i:ə/ as the form of the phoneme, since we assume that the realization of the phoneme when it precedes a dark l is influenced by the dark l. The schwa articulation, which is retracted from the high front [i:] position, is a matter of the vowel articulation assimilating to the tongue body retraction in the dark l. In doing so, we appeal to the idea of phonetic motivation: our analysis is phonetically motivated in the sense that we can provide an articulatory reason for the /i:/ → [i:ə] process, whereas we would be unable to provide any such motivation for a process in which /i:ə/ → [i:] word-finally and before any consonant other than [ɫ].

正如我们在第 5 章中所做的那样，我们还假设音位 /l/ 有两个同位异音，[l] 和 [ɫ]。我们在那儿说过有一个 /l/ 实现规则：/l/ 在元音前立即实现为 [l]，在元音后立即实现为[ɫ] 。

We are also assuming, as we did in chapter 5, that the phoneme /l/ has two allophones, [l] and [ɫ]. We said there that there is an /l/ realization rule: /l/ is realized as [l] immediately before vowels, and as [ɫ] immediately after vowels.

这两个主张似乎使我们相信 [ɫ] 出现的规则必须在某种意义上“先于”音位 /i:/ 的规则，因为我们声称 [ɫ] 只出现过作为应用 /l/ 规则的结果，[i:ə] 只会在 [ɫ] 后跟时出现。我们可以将关于这两个规则的相互作用的声明描述如下：

These two claims appear to commit us to the idea that the rule governing the occurrence of [ɫ] must, in some sense, ‘precede’ the rule governing the phoneme /i:/, since we are claiming that [ɫ] only ever arises as a result of the application of the /l/ rule, and that [i:ə] only ever arises when an [ɫ] follows. We may depict this claim about the interaction of the two rules as follows:

(15)		/fi:l/
	/l/ 规则	fi:ɫ
	/我掌管	fi:əɫ

这种描述被称为推导：音位形式 /fi:l/ 的语音实现是通过规则的有序应用从该音位形式推导出来的。因此，这种看待音系形式与其语音实现之间关系的方式使我们致力于规则排序的思想，从而使我们致力于基于规则的、派生的音系组织观点。虽然我们不会深入探讨这种语音组织的概念，但也承认我们隐含地假设了这样一个概念。

This kind of depiction is referred to as a derivation: the phonetic realization of the phonological form /fi:l/ is derived from that phonological form by means of the ordered application of rules. This way of looking at the relationship between phonological forms and their phonetic realization therefore commits us to the idea of rule ordering, and thus to a rule-based, derivational view of phonological organization. While we will not pursue this conception of phonological organization in any depth, it is as well to acknowledge that we are implicitly assuming such a conception.

英语中元音变位的另一个例子是元音长度。在许多语言中，元音长度是音位的。例如，在 Limbu（在尼泊尔使用）中，[sapma] 的意思是“写”，而 [sa:pma] 的意思是“奉承”。同样，在马拉雅拉姆语（在印度南部使用）中，[ciri] 表示“微笑”，而 [ci:ri] 表示“尖叫”。但在其他语言中，元音长度是异音。例如，在苏格兰标准英语 (SSE) 中，一些（不是全部）元音音素具有长音和短音位：

Another example of vowel allophones in English concerns vowel length. In many languages, vowel length is phonemic. For instance, in Limbu (spoken in Nepal), [sapma] means ‘to write’, but [sa:pma] means ‘to flatter’. Similarly, in Malayalam (spoken in Southern India), [ciri] means ‘smile’ but [ci:ri] means ‘shrieked’. But in other languages, vowel length is allophonic. In Scottish Standard English (SSE), for instance, some (not all) of the vowel phonemes have long and short allophones:

(16) 苏格兰标准英语中的长元音和短元音

gueq06_083

音素 /i/、/u/ 和 /ai/ 的长同位异音出现在以下上下文中：在单词末尾或以下浊辅音之一之前：/v/、/ð/、/z/ 和/ɹ/. 因此，在 SSE 中，元音长度是变位音，而不是音位。

The long allophones of the phonemes /i/, /u/ and /ai/ occur in the following contexts: at the end of a word or before one of the following voiced consonants: /v/, /ð/, /z/ and /ɹ/. Vowel length is therefore allophonic, rather than phonemic, in SSE.

gueqrecard_001

我们考虑了双唇辅音之前的辅音音素 /m/ 和 /n/ 之间的对比中和的情况。中和元音音素之间的对比也是可能的。一个例子来自GA中/æ/、/ε/和/eɪ/的对比：这三个音素在/ɹ/音素之前都被实现为[ε]。在www.wiley.com/go/carrphonetics上收听Track 6.1中的句子Marry Merry Mary。GA 说话者在这三个词中的每一个都发 [ε]：它们在 GA 中是同音字。其他说话者分别说 SSE 和 RP：每个人对三个音素中的每一个都有不同的实现：Marry Merry Mary中和不会发生在 SSE 或 RP 中。

We considered the case of the neutralization of the contrast between the consonant phonemes /m/ and /n/ before bilabial consonants. Neutralization of contrast between vowel phonemes is also possible. An example comes from the contrast between /æ/, /ε/ and /eɪ/ in GA: these three phonemes are all realized as [ε] before the /ɹ/ phoneme. Listen to the sentence Marry Merry Mary on Track 6.1 at www.wiley.com/go/carrphonetics. The GA speaker utters [ε] in each of the three words: they are homophones in GA. The other speakers are speaking SSE and RP respectively: each has distinct realizations for each of the three phonemes: the Marry Merry Mary neutralization does not occur in SSE or RP.

练习

Exercises

1 GA、SSE 和 RP 中的音位对比

以下数据集来自通用美国语（GA）、苏格兰标准英语（SSE）和RP（Received Pronunciation）。根据一个品种中存在最小对与另一个品种中不存在此类对的情况，确定 GA 中存在但 SSE 中不存在的任何音位对比，反之亦然。然后识别任何存在于 RP 中但不存在于 GA 中的音位对比，反之亦然。最后，确定 SSE 中存在的任何音位对比，但 RP 中不存在，反之亦然。假设这里给出的数据完全代表了所讨论的品种。

gueq06_084

gueq06_085

2 英语和西班牙语鼻塞

回想一下，以下数据导致我们决定假设英语中有三个鼻音塞音（/m/、/n/ 和 /ŋ/）：

英语

（一种）	[mi:t]	肉
（乙）	[ni:t]	整洁的
（C）	[moʊɫ]	痣
(四)	[否ʊɫ]	小丘
(五)	[sɪn]	罪
（F）	[sɪŋ]	唱歌
（G）	[dɪm]	暗淡
（H）	[dɪn]	喧嚣
（一世）	[wɪn]	赢
(十)	[wɪŋ]	翅膀

现在检查以下西班牙语数据中相同的三个鼻塞音（[m]、[n]、[ŋ]）。（我们忽略了西班牙语有齿音而不是牙槽鼻塞的事实。这些数据代表了卡斯蒂利亚西班牙语，这是西班牙的声望口音：许多西班牙语变体缺少 /θ/ 和 /ð/ 音位。这事实不会影响练习中提出的观点。）假设数据完全具有代表性：

西班牙语

（一种）	[muðo]	'沉默的'
（乙）	[nuðo]	'结'
（C）	[metɐ]	'目标'
(四)	[网络]	'纯'（女性）
(五)	[ɔmbɾe]	'掌管'
（F）	[sεmblɐɾ]	'显得'
（G）	[ɐndɐɾ]	'去'
（H）	[ɐnte]	“面对”
（一世）	[ɐŋgulo]	'角度'
(十)	[lεŋgwɐ]	'语'

注意。[ŋ] 不出现在词首、词尾或元音之间。

（一世）	有什么证据可以假设西班牙语中有不同的 /m/ 和 /n/ 音位？
(二)	是否有任何理由假设西班牙语中有 /ŋ/ 音位？你的回答有什么证据？
现在看看下面的西班牙数据：

(k)	[feliθ]	'快乐的'
(l)	[lɱfeliθ]	'不开心'
(米)	[可能]	'可能的'
(n)	[不可能]	'不可能的'
(o)	[可有可无]	'可有可无'
(p)	[不可或缺]	“不可或缺”
(q)	[阅读]	'可以想象'
(r)	[iŋkonθebible]	'不可思议'
（小）	[ɐkɔstɔmbɾɐðo]	'习惯'
(t)	[inɐkɔstɔmbɾɐðo]	'不习惯'

(三)

西班牙鼻音塞音是否有任何音位重叠？解释。

gueqrecard_002

3 进一步的音标练习

听曲目 6.2。尽可能详细地转录您在录音中听到的单词。话是：

不友好的完整的独立芦苇卷轴

7 英语音节结构

English Syllable Structure

7.1 简介

7.1 Introduction

我们已经说过语素是一种心理表征，它具有三个属性：句法范畴、意义和音位形式。到目前为止，我们已经允许语素的音位形式出现在说话者心理构成的语法中，并且这种音位形式包含在单个音位片段或这样的片段序列中。但这只是故事的一部分：语素的音韵形式远不止于此。有证据表明，这些片段被组织成语音成分，而不是单词被组织成句法成分（例如短语和句子）的方式。这些成分之一是音节. 音节存在的证据主要以语音概括的形式出现，如果不参考“音节”的概念就无法充分表达。本章的目的是检查英语音节的结构，并举例说明用该结构最好地表达的一些音位概括。

We have said morphemes are a kind of mental representation which have three properties: a syntactic category, a meaning and a phonological form. We have allowed, thus far, that the phonological form of a morpheme is present in the speaker’s mentally constituted grammar, and that this phonological form consists in either a single phonological segment or a sequence of such segments. But this is only part of the story: there is more to the phonological form of a morpheme than that. There is evidence that those segments are organized into phonological constituents, rather in the way that words are organized into syntactic constituents (such as phrases and sentences). One of those constituents is the syllable. The evidence for the existence of the syllable comes largely in the form of phonological generalizations which cannot be adequately expressed without reference to the notion ‘syllable’. The aim of this chapter is to examine the structure of the syllable in English, and exemplify some of the sorts of phonological generalization which are best expressed in terms of that structure.

7.2 音节结构中的选区

7.2 Constituency in Syllable Structure

音节中的两个主要成分是首字母和押韵。例如，在单词bile中，第一段 /b/ 构成音节的开头，最后两段 /aɪ/ 和 /l/ 合在一起构成韵脚。首字母定义为出现在元音之前的任何和所有辅音。有什么证据可以区分起首和押韵？头韵的装置取决于词首的身份，独立于押韵的内容，如little and light , poor and packed，等等。这构成了首字母/韵律划分的证据，从而证明押韵是一个有充分根据的音节成分。既然如此，那么开始作为一个组成部分同样是有根据的（因为两者的定义是相互矛盾的）。口误也表明起音是言语产生中的一个真实单元。一种口误是 spoonerism，以一位名叫 Spooner 的学者的名字命名，据说他说出诸如“你错过了我的历史讲座”之类的句子是“你已经听到了我的神秘讲座”，并且将“错过”和“历史”的开始。

The two main constituents within a syllable are the onset and the rhyme. In the word bile, for instance, the first segment, /b/, constitutes the onset of the syllable and the last two segments, /aɪ/ and /l/, taken together, constitute the rhyme. The onset is defined as any and all consonants occurring before the vowel. What evidence is there for this division between onset and rhyme? The device of alliteration depends on identity of onsets, independently of the content of the rhyme, as in little and light, poor and packed, and so on. This constitutes evidence for the onset/rhyme division, and thus evidence that the rhyme is a well-founded syllabic constituent. Since that is so, then the onset as a constituent is equally well-founded (since the two are defined in contradistinction to each other). Slips of the tongue also show that the onset is a real unit in speech production. One type of slip of the tongue is the spoonerism, named after an academic called Spooner, who is said to have uttered sentences such as ‘You have missed my history lecture’ as ‘You have hissed my mystery lecture’, with an inversion of the onsets of ‘missed’ and ‘history’.

押韵可以进一步细分为成分核心和尾声。因此，在单词bile中，双元音 /aɪ/ 构成核心，辅音 /l/ 构成尾音。我们可以这样表示单音节语素胆的选区，其中希腊语‘σ’（sigma）代表‘音节’，‘O’代表‘开头’，‘R’代表‘押韵’，‘N’代表“核”，“C”代表“结尾”：

The rhyme may be further subdivided into the constituents nucleus and coda. Thus, in the word bile, the diphthong /aɪ/ constitutes the nucleus, and the consonant /l/ constitutes the coda. We may represent the constituency of the single-syllable morpheme bile as follows, where Greek ‘σ’ (sigma) stands for ‘syllable’, ‘O’ stands for ‘onset’, ‘R’ stands for ‘rhyme’, ‘N’ stands for ‘nucleus’, and ‘C’ stands for ‘coda’:

(1)胆汁：

像这样的音节在尾音位置包含一个或多个辅音，称为闭音节，而在尾音位置不包含任何辅音的音节称为开音节；就像买这个词一样：

A syllable such as this, which contains one or more consonants in coda position, is called a closed syllable, whereas a syllable which does not contain any consonants in coda position is referred to as an open syllable; as in the word buy:

(2)购买：

虽然一个音节必须有一个核心，但有可能有一个结构良好的音节，它不包含核心以外的任何元素。占据音节核心的部分通常是元音。一个词的例子在只有一个音节的英语中，它又只包含一个核心，是眼睛：/aɪ/。但是，正如我们所见，英语音节的核心可能位于其他部分之前或之后，而这些部分通常是辅音。例如，在单词aisle中，核心后跟一个尾音位置的辅音：/aɪl/。在单词buy中，核前有一个位于声母位置的辅音：/baɪ/，而在单词bile中，核前和后都有辅音：/baɪl/。

While a syllable must have a nucleus, it is possible to have a well-formed syllable which does not contain any element other than a nucleus. The segment occupying the nucleus of the syllable is normally a vowel. An example of a word in English consisting of only one syllable, which in turn contains only a nucleus, is eye: /aɪ/. But the nucleus of a syllable in English may be preceded or followed by other segments, as we have seen, and those segments are typically consonants. In the word aisle, for instance, the nucleus is followed by a consonant in coda position: /aɪl/. In the word buy, the nucleus is preceded by a consonant in onset position: /baɪ/, and in the word bile, the nucleus is both preceded and followed by consonants: /baɪl/.

像bile这样只包含一个音节的词素被称为单音节词素。在某些语言中，所有语素都是单音节的。但在英语中，语素可能包含不止一个音节：它们可能是多音节的。例如rider、beetle、amount、desire（双音节）、elephant、veranda、kangaroo（三音节）、independent、America（四音节）等。

Morphemes like bile, which contain only one syllable, are said to be monosyllabic. In some languages, all morphemes are monosyllabic. But in English, morphemes may contain more than one syllable: they may be polysyllabic. Examples are rider, beetle, amount, desire (which are bisyllabic), elephant, veranda, kangaroo (which are trisyllabic), independent, America (which have four syllables) and so on.

在某些语言中，所有音节都必须包含一个起始辅音，但正如我们所见，在英语中（许多其他语言也是如此），情况并非如此。出于稍后解释的原因（与“重新音节化”的概念相关），我们将用空开头表示此类音节，如下所示：

In some languages, all syllables must contain an onset consonant but, as we have seen, in English (and this is true of many other languages), this is not the case. For reasons to be explained later (connected with the notion of ‘resyllabification’), we will represent such syllables with an empty onset, as follows:

(3)它：

在许多语言中，例如夏威夷语，词首可能只包含一个辅音，但在许多其他语言中，包括英语，词首可能包含两个部分（如bring、trap、clip等）；我们将这些称为branching onsets，并将它们表示如下：

In many languages, such as Hawaiian, onsets may contain a single consonant only, but in many others, English included, onsets may contain two segments (as in bring, trap, clip, etc.); we will refer to these as branching onsets, and represent them as follows:

(4)剪辑：

正如开头可能有分支一样，尾声也可能有分支，如hunt一词：

Just as onsets may be branching, so codas may branch, as in the word hunt:

(5)打猎：

一方面，短元音与另一方面，长元音和双元音之间的区别可以通过后一类元音占据一个分支核，而前一类占据一个非分支核来表示。为了表示长元音和双元音比短元音长的事实，我们说片段附加到一系列时隙，称为骨架层。这个想法是一方面可以表示短元音，另一方面可以表示长元音（包括英语双元音）之间的区别，方法是将前者连接到单个骨架槽，将后者连接到两个骨架槽插槽。因此，位有一个短元音作为它的核，因此用一个非分支核表示，而bee和buy有分支核：

The distinction between, on the one hand, short vowels and, on the other, long vowels and diphthongs can be represented by taking the latter class of vowels to occupy a branching nucleus, with the former class occupying a non-branching nucleus. To represent the fact that long vowels and diphthongs are longer than short vowels, we say that segments are attached to a series of timing slots, referred to as the skeletal tier. The idea is that one can represent the difference between short vowels on the one hand, and long vowels (including English diphthongs) on the other, by taking the former to be connected to a single skeletal slot and the latter to be connected to two skeletal slots. Thus, bit has a short vowel as its nucleus and is therefore represented with a non-branching nucleus, whereas bee and buy have branching nuclei:

(6)位：

(7)蜜蜂：

(8)购买：

(7) 和 (8) 中的表示的意图是，长元音被构成为附加到两个骨架槽的单个元音质量，而长双元音，如buy中，具有两种不同的元音质量。关键在于，具有长元音和双元音的核与核内的时隙数量是平行的。今后我们将在音节结构的表示中采用骨架层。

What is intended by the representations in (7) and (8) is that long vowels are constituted as a single vowel quality which is attached to two skeletal slots, whereas long diphthongs, as in buy, have two different vowel qualities. The point is that nuclei with long vowels and with diphthongs are parallel with respect to the number of timing slots within the nucleus. We will henceforth adopt the skeletal tier in our representations of syllabic structure.

骨架层使我们能够说擦擦音，正如我们所见，有一个闭合元素和一个擦音释放元素，如 [tʃ] 和 [dʒ]，是复杂的片段，因为它们表现得像单个片段（它们占据单个时间单位）同时具有类似于两个部分的内部结构：

The skeletal tier enables us to say that affricates, which, as we have seen, have a closure element and a fricative release element, as in [tʃ] and [dʒ], are complex segments, since they behave like single segments (they occupy a single unit of timing) while having an internal structure which resembles two segments:

(9)芯片：

7.3 响度等级、最大声母和音节权重

7.3 The Sonority Hierarchy, Maximal Onset and Syllable Weight

如果你的母语是英语，你会同意 /bliŋk/ 是一个结构良好的音节（并且恰好构成英语单词的音韵形式），如下所示：/blʊŋk/, /blεŋk/, /blɐŋk/ 和 /blæŋk/（其中大部分碰巧不构成英语单词的音韵形式）。也就是说，您的母语人士的英语知识可以让您判断这些词在音节上是正确的，即使英语中没有单词具有这些音韵形式。这种无意识的知识还可以让你判断以下是错误格式的：/ibɪŋk/、/ibɪkŋ/、/tlɪnk/ 和 /blaɪmp/。问题是：这种无意识知识采取什么形式？我们不知不觉中知道的是什么让我们做出这些判断？现在让我们设法回答这个问题。

You will agree, if you are a native speaker of English, that /bliŋk/ is a well-formed syllable (and happens to constitute the phonological form of an English word), as are the following: /blʊŋk/, /blεŋk/, /blɐŋk/ and /blæŋk/ (most of which happen not to constitute the phonological form of English words). That is, your native speaker knowledge of English allows you to judge that these are syllabically well-formed, even though there are no words in English which have those phonological forms. That unconscious knowledge also allows you to judge that the following are ill-formed: /ibɪŋk/, /ibɪkŋ/, /tlɪnk/ and /blaɪmp/. The question is: what form does this unconscious knowledge take? What is it that we know, unconsciously, which allows us to make these judgements? Let us now seek to answer that question.

人们普遍认为，对音节可能采用的形式存在通用和语言特定的限制，即对片段序列的音节化的限制。在通用约束中，我们可以提到两个。首先，据称片段序列是根据响度等级进行音节化的，其形式如下：

It is widely believed that there are both universal and language-specific constraints on the form that syllables may take, that is, constraints on the syllabification of sequences of segments. Among the universal constraints, we may mention two. Firstly, it is claimed that sequences of segments are syllabified in accordance with a sonority scale, which takes the following form:

(10) 响度等级

低元音

高元音

近似值

鼻音

浊擦音

清擦音

浊音停止

无声停止

这个想法是，当一个人从音阶的底部到顶部前进时，音段的类别变得更响亮，或更像元音。响度是一种声学效果：声音越响亮，共鸣越多。元音比辅音有更大的共鸣，浊辅音比清辅音有更大的共鸣。如果您长时间聆听一位歌手按住一个音符，所讨论的声音很可能是元音。元音比辅音更容易保持更长时间，这有两个发音原因，两者都与响度的产生有关。首先是收缩程度，正如第 1 章中所讨论的那样，据称塞音比擦音包含更大程度的收缩，而摩擦音又包含比逼近音和元音更大的收缩。同样，元音发音越开放，口腔中的狭窄就越少。这些发音的声学效果是收缩程度越小，响度越大。第二清晰度因素是浊音：清音片段不像浊音片段那样像元音，也不太响亮：元音通常是浊音，浊音会产生更大的响度。

The idea is that, as one proceeds from the bottom to the top of the scale, the class of segments becomes more sonorous, or more vowel-like. Sonority is an acoustic effect: the more sonorous a sound, the more it resonates. Vowels have greater resonance than consonants, and voiced consonants have greater resonance than voiceless ones. If you listen to a singer holding a note for any length of time, the sound in question will most probably be a vowel. There are two articulatory reasons why it is easier to hold a vowel sound for longer than a consonant sound, and both are relevant to the production of sonority. The first is degree of constriction, as discussed in chapter 1, whereby stops are said to involve a greater degree of stricture than fricatives, which in turn involve greater constriction than approximants and vowels. Similarly, the more open a vowel articulation, the less stricture there is in the oral cavity. The acoustic effect of these sorts of articulation is that the lesser the degree of constriction, the greater the degree of sonority. The second articulatory factor is voicing: voiceless segments are less vowel-like, less sonorant, than voiced segments: vowels are typically voiced, and voicing creates greater sonority.

应用于音节结构，这个想法是音节中最响亮的元素将位于核心内，并且离核心越远的音节就越不响亮。因此，在blink中，/b/ 不如 /l/ 响亮，而 /l/ 又不如元音响亮：当一个人接近核心时，响度会增加。当离开核心时，我们可能会注意到 /ŋ/ 不如元音响亮，而 /k/ 又不如前面的 /ŋ/ 响亮。

Applied to syllable structure, the idea is that the most sonorous element in a syllable will be located within the nucleus, and that the further one gets from the nucleus, the less sonorous are the segments. Thus, in blink, the /b/ is less sonorous than the /l/, which is, in turn, less sonorous than the vowel: as one approaches the nucleus, so sonority increases. As one leaves the nucleus, we may note that the /ŋ/ is less sonorous than the vowel, and the /k/ less sonorous in turn than the preceding /ŋ/.

“响亮程度”的想法非常有说服力，即使它遇到了一些困难。例如，英语中的 [s] + 辅音起始簇破坏了响度等级所做的预测，因为在诸如sprint之类的情况下，除了初始 [s] 之外，响度等级原则做出了正确的预测。然而，这只是进一步强调了英语 sC (s + 辅音) 起始音簇的特殊性：只有 [s]-初始起始音违反了响度等级，英语中唯一的三向分支起始音是那些以 [秒]。

The ‘degree of sonority’ idea is very convincing, even if it runs into some difficulties. For instance, [s] + consonant onset clusters in English undermine the predictions made by the sonority hierarchy, since, in cases such as sprint, the sonority scale principle makes the right predictions except with respect to the initial [s]. However, this merely serves further to underline the peculiarity of English sC (s + consonant) onset clusters: only [s]-initial onsets violate the sonority hierarchy, and the only three-way branching onsets in English are those which begin with an [s].

音节化的另一个普遍原则是关于多音节词的音节化，被称为最大声母原则。到目前为止，我们只考虑了单音节词；因此，让我们考虑英语单词 appraise 的音节化，假设其音节形式为 / əpɹeɪz /。很明显，这个词是双音节的；问题是音节之间的边界在哪里。我们知道 /p/ 可能出现在英语的结尾位置，如cap、cup等。我们也知道 /pɹ/ 是一个格式正确的词首，如prize、preen等，并且我们知道 / ɹ/ 可能单独出现在起始位置，如在rice、raze中等等。此外，我们知道 /pɹ/ 不是一个格式正确的尾音簇：它违反了响度等级的预测。因此，/u:pɹ/、/sɪpɹ/ 等格式不正确。因此，我们必须确定 appraise 的音节化是 /ə.pɹeɪz/ 还是/əp.ɹeɪz/（句号表示音节边界）。Maximal Onset 的原则说，在这种情况下，语言特定的语音策略将允许跨越音节边界的两个或更多音节化，这是首选的音节化，它最大化了随后的开始中的材料。在这种情况下，这是前一个音节。

Another universal principle of syllabification concerns the syllabification of polysyllabic words, and is referred to as the principle of Maximal Onset. We have considered only monosyllabic words thus far; let us therefore consider the syllabification of the English word appraise, whose segmental form is, let us say, /əpɹeɪz/. It is clear that the word is bisyllabic; the question is where the boundary between the syllables lies. We know that /p/ may occur in coda position in English, as in cap, cup, etc. We also know that /pɹ/ is a well-formed onset, as in prize, preen, etc., and we know that /ɹ/ may occur alone in onset position, as in rice, raze, etc. Furthermore, we know that /pɹ/ is not a well-formed coda cluster: it violates the predictions of the sonority hierarchy. Thus, /u:pɹ/, /sɪpɹ/ etc. are ill-formed. We must therefore decide whether the syllabification of appraise is /ə.pɹeɪz/ or /əp.ɹeɪz/ (where the full stop indicates the syllable boundary). The principle of Maximal Onset says that, in cases like this, where the language-specific phonotactics will allow for two or more syllabifications across a syllable boundary, it is the syllabification which maximizes the material in the following onset which is preferred. In this case, that is the former syllabification.

最大声母原则与关于音节结构的一个普遍事实密切相关：具有声母辅音的音节在某种意义上比没有声母的音节更基本，并且存在声母辅音在某种意义上比存在尾辅音更基本。人类语言中最“基本”的音节结构似乎是 CV 音节结构，单个起始辅音后跟一个元音。这种说法有几种类型的证据。

The principle of Maximal Onset is intimately connected with a universal fact about syllable structure: that syllables with an onset consonant are in some sense more basic than those without, and that presence of onset consonants is in some sense more basic than presence of coda consonants. It appears that the most ‘basic’ syllable structure in human languages is CV syllable structure, with a single onset consonant followed by a vowel. There are several types of evidence for this claim.

首先，CV 型音节似乎是人类儿童在开始说话时首先说出的音节类型（例如 [ba]、[ma]），而不管他们的父母说什么语言。在儿童音节结构发展的那个阶段，成人语言中带有分支首字母的音节将作为 CV 结构说出来。带有尾辅音的音节也是如此：尾辅音将只是在那个阶段缺席。这有力地表明，起始辅音在某种程度上比尾辅音更基本，在发音方面，也许在感知方面。

Firstly, CV-type syllables appear to be the syllable types that human children first utter when they begin to speak (e.g. [ba], [ma]) regardless of what language their parents speak. At that stage in the development of the child’s syllable structure, syllables in the adult language with branching onsets will be uttered as CV structures. So too will syllables with coda consonants: the coda consonants will simply be absent at that stage. This strongly suggests that onset consonants are in some way more basic, in articulatory, and perhaps perceptual, terms than coda consonants.

其次，在许多失语症病例中，中风后患者的言语受到损害，CV 音节结构似乎也是患者恢复言语时最先出现的那种，即使他或她的母语有分支起始和尾辅音。

Secondly, in many cases of aphasia, where post-stroke patients have suffered damage to their speech, CV syllable structures also appear to be the sort that first begins to appear as the patient recovers his or her speech, even if his or her native language has branching onsets and coda consonants.

第三，同时具有起始和尾声辅音的语言通常允许更广泛的辅音出现在起始位置而不是尾声位置。

Thirdly, languages which have both onset and coda consonants typically allow for a wider range of consonants to occur in onset position than in coda position.

第四，在语言的历史发展过程中，尾辅音比起首辅音更容易发生发音损失。这就是某些英语变体中 /l/ 在尾音位置发生的情况，其中 /l/ 的实现在尾音位置变成发声（[w]，类似于元音，而不是辅音），但不是起始位置，因此 [l] 出现在像let和play这样的词中，其中 /l/ 处于起始位置，但 [w] 出现在像feel和felt等词中，其中 /l/ 出现在结尾位置（除了感觉等词的情况后跟以空首字母开头的单词或后缀，在这种情况下 /l/ 占据该位置并实现为 [l]；参见下面关于重新音节化的 7.7）。这种发音减弱会导致辅音完全省略（不发音）。这就是许多英语口音中 [ɹ] 在结尾位置发生的情况。在这些口音中，像car和card这样的词去掉了词尾 [ɹ]，但在词首保留了它，例如run和bring等词。

Fourthly, coda consonants are much more likely to undergo loss of articulation in the course of the historical development of languages than onset consonants. This is what has happened with /l/ in coda position in some varieties of English, where the realization of /l/ has become vocalized ([w], which is vowel-like, rather than consonantal) in coda position, but not in onset position, so that [l] occurs in words like let and play, where the /l/ is in an onset, but [w] occurs in words like feel and felt, where the /l/ occurs in coda position (except in cases where words such as feel are followed by a word or suffix beginning with an empty onset, in which case the /l/ occupies that position and is realized as [l]; see 7.7 below on resyllabification). This kind of weakening of articulation can lead to complete elision (non-pronunciation) of a consonant. This is what has happened with [ɹ] in coda position in many accents of English. In those accents, words like car and card have lost the coda [ɹ], while retaining it in onsets, as in words like run and bring.

这种发音减弱的情况，通常导致完全丧失发音，尾辅音在世界语言中比比皆是。他们认为尾辅音在某种程度上不如起始辅音在感知上显着，并且对人类从心理存储中检索音位形式的方式的研究表明起始辅音比尾辅音更突出：如果一个人在一个人的词汇中搜索一个词记忆，人们更有可能根据起始辅音而不是尾辅音进行搜索。

Such cases of articulatory weakening, often leading to complete loss of articulation, of coda consonants abound in the world’s languages. They suggest that coda consonants are somehow less salient in perception than onset consonants, and studies in the way that human beings retrieve phonological forms from mental storage suggest greater prominence for onset consonants than for coda consonants: if one is searching for a word in one’s lexical memory, one is more likely to search on the basis of onset consonants than of coda consonants.

第五，没有已知的语言有VC型音节而没有CV型音节，反之则不然。这强烈表明 CV 音节比 VC 或任何其他音节类型更基本。

Fifthly, there are no known languages which have VC-type syllables but lack CV-type syllables, whereas the reverse is not the case. This strongly suggests that CV syllables are more basic than VC, or indeed any other, syllable type.

这种关于 CV 音节结构的概括可能在发音和感知方面都有基础。如果你试图独立地产生一个空开头的词（例如词eye），你会发现在你发出元音之前不发出某种辅音发音（通常是声门塞音）是很难做到的。对充满而不是空声的偏好可能植根于我们发音器官的性质，也与声母的感知显着性有关。

This generalization about CV syllable structure probably has a basis in both articulation and perception. If you try to produce a word with an empty onset in isolation (e.g. the word eye), you will find it hard to do without uttering some kind of consonantal articulation (typically, a glottal stop) before you utter the vowel. Preference for filled, rather than empty, onsets is probably rooted in the nature of our articulatory apparatus and also tied to greater perceptual salience of onset consonants.

鉴于最大起音的原则，很明显，在诸如 appraise 中的第一个音节之类的音节中，押韵包含一个短元音（当然，由单个骨架槽支配）并且不包含尾音，因此：

Given the principle of Maximal Onset, it is clear that, in a syllable such as the first syllable in appraise, the rhyme contains a short vowel (dominated, of course, by a single skeletal slot) and does not contain a coda, thus:

(11)评价：

诸如 appraise 中的第一个音节之类的音节，无论是在韵节本身的层次上，还是在韵核中，都没有在韵脚内分支，因此被称为轻音节。在韵律成分中任何地方都有分支的音节被称为重音节。一些人认为这种音节重量的区别对于理解英语单词重音的性质很重要。

Syllables such as the first syllable in appraise, in which there is no branching within the rhyme, either at the level of the rhyme node itself, or within the nucleus, are called light syllables. And syllables which have branching anywhere within the rhyme constituent are called heavy syllables. This distinction in syllable weight is said by some to be important in understanding the nature of word stress in English.

一些音韵学家对英语中的单词重音有两种概括。首先是英语中任何重读音节很可能是重音节。第二个是单音节词可能不会以短元音音素之一（/ʊ/、/ɪ/、/æ/、/ε/、/ɑ/ 或 /ʌ/）结尾，因为一个核心仅包含其中一个音素没有尾辅音的元音很轻，如果要重读一个单音节词，则无法选择重读哪个音节。

There are two generalizations about word stress in English which some phonologists make. The first is that any stressed syllable in English is very likely to be a heavy syllable. The second is that monosyllabic words may not end in one of the short vowel phonemes (/ʊ/, /ɪ/, /æ/, /ε/, /ɑ/ or /ʌ/), since a nucleus containing only one of those vowels, with no coda consonant, is light, and if a monosyllabic word is to be stressed, there is no choice as to which syllable it will be stressed on.

7.4 特定语言的语音策略

7.4 Language-Specific Phonotactics

现在让我们考虑一些特定于语言的对片段序列的限制，这些限制可以组合在音节结构中，称为特定于语言的语音限制（简称语音）。

Let us now consider some language-specific constraints on the sequences of segments which may be combined in syllable structure, known as language-specific phonotactic constraints (phonotactics, for short).

我们已经允许，在英语音节中，首字母、核心、押韵和结尾可以分支。但我们没有说他们可能拥有的分支机构数量是否有任何限制。只有一种英语首字母超过二元（双向）分支：/s/ + 辅音 + {/j/、/w/ 或 /ɹ/} 首字母，如spew、square和scream。请注意，可能构成此类序列中第三个元素的片段范围甚至比二进制分支起始中的片段范围更受限制。

We have allowed that, in English syllables, onsets, nuclei, rhymes and codas may branch. But we have not said whether there is any limit on the number of branches they may have. Only one sort of English onset exceeds binary (two-way) branching: /s/ + consonant + {/j/, /w/ or /ɹ/} onsets, as in spew, square and scream. Note that the range of segments which may form the third element in such sequences is even more restricted than those in binary branching onsets.

正如我们所见，英语中的首字母可能会发生分支，但如果发生分支，则它们可能采用的形式存在音序限制。忽略 /s/ + 辅音的情况，我们可以说第一段必须是塞音或擦音，第二段必须是 /ɹ/、/l/、/j/ 或 /w/。因此 /pɹ/, /pl/, /pj/, /bɹ/, /bl/, /bj/, /tɹ/, /tw/, /dɹ/, /dw/, /kɹ/, /kl/, / kw/、/θɹ/、/θw/、/fɹ/、/fl/、/fj/、/sl/、/sj/ 和 /sw/ 都是允许的。此列表反映了其他开始的语音策略。例如，/t/、/d/ 和 /θ/ 后面可能没有 /l/，并且没有一个浊擦音可能出现在分支声母中。

As we have seen, onsets may branch in English, but if they do, there are phonotactic constraints on the form they may take. Ignoring the /s/ + consonant cases, we may say that the first segment must be a stop or a fricative and the second must be /ɹ/, /l/, /j/ or /w/. Thus /pɹ/, /pl/, /pj/, /bɹ/, /bl/, /bj/, /tɹ/, /tw/, /dɹ/, /dw/, /kɹ/, /kl/, /kw/, /θɹ/, /θw/, /fɹ/, /fl/, /fj/, /sl/, /sj/ and /sw/ are all permissible. This list reflects other onset phonotactics. For instance, /t/, /d/ and /θ/ may not be followed by /l/, and none of the voiced fricatives may occur in branching onsets.

在英语押韵的语音限制中，我们可能会注意到以下几点。首先，/h/ 不出现在英文押韵中。其次，在许多英语口音中，/ɹ/ 也不出现在押韵中；所以像farm和car这样的词可以说有 /fɑ:m/ 和 /kɑ:/ 这样的音韵形式，没有 /ɹ/。押韵中缺少 /ɹ/ 的口音被称为非流音口音；它们包括澳大利亚英语、新西兰英语、RP、南非英语、英格兰北部的大部分口音以及美国南部和东部的口音。这些口音在某个阶段很吵；[ɹ] 在这些口音中丢失了押韵。没有经历过这种历史变化的 Rhotic accents 包括 GA，在苏格兰使用的英语口音，以及在英格兰西南部使用的一些口音。稍后我们将更详细地讨论这些口音。

Among the phonotactic constraints on rhymes in English, we may note the following. Firstly, /h/ does not occur in rhymes in English. Secondly, in many accents of English, /ɹ/ does not occur in rhymes either; so that words like farm and car arguably have phonological forms such as /fɑ:m/ and /kɑ:/, without an /ɹ/. Accents which lack /ɹ/ in rhymes are referred to as non-rhotic accents; they include Australian English, New Zealand English, RP, South African English, most of the accents of the North of England, and the Southern and Eastern accents of the United States. These accents were rhotic at one stage; [ɹ] has been lost in rhymes in those accents. Rhotic accents, which have not undergone this historical change, include GA, the accents of English spoken in Scotland, and some accents spoken in the South West of England. We will discuss such accents in more detail later.

一种语言中音节的整体形状通常是成人第二语言习得的主要因素。例如，稍微简化一下，日语音节结构不允许分支起始。这通常会产生这样的效果，即当以日语为母语的人说出具有复杂首字母的英语单词时，例如screw，他们往往会在前三个辅音中的每一个之后插入一个元音，从而使单词成为三音节：[sɯkɯɾɯ]。这种插入元音的过程称为元音插音，此类元音称为插音元音。同样，再次稍微简化，日语音节结构不允许词尾辅音，因此英语借词，如蛋糕, 在英语中以尾音辅音结尾，通常发音为以元音结尾的双音节词：[keki]。

The overall shape of syllables in a language often acts as a major factor in adult second language acquisition. For instance, simplifying somewhat, Japanese syllable structure does not allow for branching onsets. This often has the effect that, when native speakers of Japanese utter English words with complex onsets, such as screw, they will tend to insert a vowel after each of the first three consonants, thus rendering the word trisyllabic: [sɯkɯɾɯ]. This process of vowel insertion is known as vowel epenthesis, and such vowels are known as epenthetic vowels. Similarly, and again simplifying somewhat, Japanese syllable structure does not allow for word-final coda consonants, so that English loanwords, such as cake, which end in a coda consonant in English, tend to be uttered as bisyllabic words ending in a vowel: [keki].

类似的案例比比皆是。例如，与英语不同，西班牙语没有以 s + 辅音开头的单词。然而，西班牙语确实有诸如España之类的词，在某些情况下，它对应于以 s + 辅音开头的英语单词（在本例中为Spain一词）。其影响之一是说西班牙语的人倾向于在以 s + 辅音簇开头的英语单词前插入 [ε]，如 [εspeɪn]（西班牙）。类似地，西班牙语中的英语外来词用这样的附注 [ε] 发音，如 [εsmokin]（'吸烟夹克'）。

Similar cases abound. For instance, Spanish, unlike English, does not have words beginning with an s + consonant onset. However, Spanish does have words such as España which, in some cases, correspond to English words with an s + consonant onset (in this case, the word Spain). One of the effects of this is that Spanish speakers tend to insert an [ε] before English words beginning with s + consonant clusters, as in [εspeɪn] (Spain). Similarly, English loanwords in Spanish are pronounced with such an epenthetic [ε], as in [εsmokin] (‘smoking jacket’).

7.5 音节辅音和音韵

7.5 Syllabic Consonants and Phonotactics

辅音有可能在说英语的人说话时构成音节的核心，尤其是当语速增加时。这些辅音称为音节辅音。对于许多人来说， bottle这个词的三个替代发音是 [bɑtəə]、[bɑtɫ̩] 或 [bɑʔɫ̩]。在后两个发音中，最后的非重读元音 (schwa) 已经丢失，但这个词仍然有两个音节，与侧音变音节化。音节辅音通过“音节”变音符号转录，置于适当的辅音符号之下。

It is possible for consonants to form the nucleus of a syllable in the speech of English speakers, particularly as the rate of speech increases. These consonants are called syllabic consonants. Three alternative pronunciations of the word bottle, for many speakers, are [bɑtəɫ], [bɑtɫ̩] or [bɑʔɫ̩]. In the latter two pronunciations, the final unstressed vowel (schwa) has been lost, but the word still has two syllables, with the lateral becoming syllabified. Syllabic consonants are transcribed by means of the ‘syllabic’ diacritic, placed under the appropriate consonant symbol.

音节鼻音在许多英语变体中都很常见。一个例子是单词button，它有两个音节。对于许多说英语的人来说，它可能发音为 [bʌtən] 或 [bʌʔn̩]，其中第二个发音带有音节鼻音。第一个发音中的第二个元音是一个非重读元音 (schwa)，它可能会“丢失”，尤其是在更快或更随意的演讲中。一个类似的例子是单词happen，对于许多说话者来说，它（至少）有两个发音 [hæpən] 和 [hæʔm̩]。在这里，鼻音 [n] 同化为“预期的”双唇发音 [p]，后者又发音为声门塞音。其他涉及鼻音的例子有 [ɹεʔŋ ] 与 [ɹεkən] ( reckon )和 [kn̩u:] 与 [kənu:] ( canoe).

Syllabic nasals are common in many varieties of English. An example is the word button, which has two syllables. For many speakers of English, it may be pronounced [bʌtən] or [bʌʔn̩], where the second pronunciation has a syllabic nasal. The second vowel in the first pronunciation is an unstressed vowel (schwa) which may be ‘lost’, particularly in faster or more casual speech. A similar example is the word happen, which, for many speakers, has (at least) the two pronunciations [hæpən] and [hæʔm̩]. Here, the nasal [n] assimilates to the ‘intended’ bilabial articulation [p], which in turn is articulated as a glottal stop. Other examples involving nasals are [ɹεʔŋ̩] vs [ɹεkən] (reckon), and [kn̩u:] vs [kənu:] (canoe).

一个涉及近似词 [ɹ] 的类似例子是单词parade，它通常有替代发音 [pəɹeɪd] 和 [pɹ̩eɪd]。在英语中，通常是鼻音、侧音和 [ɹ] 进行音节化，尽管擦音也可能被音化，例如在某些发音中，support可能是 [səpʰɔ:t] 或 [s̩pʰɔ:t]（如不同于sport : [spɔ:t]，只有一个音节，而不是两个音节）。

A similar example involving the approximant [ɹ] is the word parade, which often has the alternative pronunciations [pəɹeɪd] and [pɹ̩eɪd]. It is typically nasals, laterals and [ɹ] which undergo syllabification in English, although fricatives may also be syllabified, as in some pronunciations of, for instance, support, which may be [səpʰɔ:t] or [s̩pʰɔ:t] (as distinct from sport: [spɔ:t], with one, not two, syllables).

在英语中（但不是在某些其他语言中），对于可能出现音节辅音的每种情况，在音节化辅音之前或之后都有元音的单词的替代发音。

In English (but not in some other languages), for every case in which a syllabic consonant may occur, there will be an alternative pronunciation of the word with a vowel preceding or following the syllabified consonant.

所有这些词都有在每个音节的核心中包含一个元音的音位表示，例如在canoe中：

All of these words have phonological representations containing a vowel in the nucleus of each syllable, as in canoe:

(12)独木舟：

令人惊讶的是，虽然说英语的人经常说出诸如[kn̩u:] 中带有音节鼻音的canoe之类的词，但当面对诸如gnu之类的非英语词时，他们往往会插入一个附音元音，将这个词说成[gənu]，使其符合英语音节结构，其中/gn̩/不是允许的英语音节。这里要指出的重要一点是，对英语音节结构的限制是根据允许的音位序列而不是语音序列来定义的。

It is striking that, although English speakers frequently utter words such as canoe with a syllabic nasal, as in [kn̩u:], when faced with non-English words such as gnu, they will tend to insert an epenthetic vowel, uttering the word as [gənu], making it conform to English syllable structure, in which /gn̩/ is not a permissible English syllable. The important point to be made here is that constraints on English syllable structure are defined in terms of permissible phonological sequences, rather than phonetic ones.

我们说过英语不允许音韵辅音，但许多其他语言允许。例如，波利尼西亚语毛利语和许多班图语都允许音韵鼻音。例如班图人的名字，例如Mbabande、Ndola和Nkomo，每个名字在语音上都以音节鼻音开头，如 [ŋ _]komo]，它有三个音节。以英语为母语的人在说出此类词时，往往会在相关鼻音附近加上附音元音，例如 [nɪkomo] 或 [ɪŋkomo]，从而使语音序列符合英语的音韵模式。所有这些案例都为我们得出的语音与语音区别提供了证据，并显示了我们的语音表征对我们对非母语单词的感知和产生有多么深远的影响。

We have said that English does not allow for phonological syllabic consonants, but many other languages do. The Polynesian language Maori, and many Bantu languages, for instance, allow for phonologically syllabic nasals. Examples are Bantu names such as Mbabande, Ndola and Nkomo, each of which begins, phonologically, with a syllabic nasal, as in [ŋ̩komo], which has three syllables. Native speakers of English will tend to utter such words with an epenthetic vowel placed adjacent to the relevant nasal, as in [nɪkomo] or [ɪŋkomo], thus making the phonetic sequence conform to the English phonological pattern. All of these cases provide evidence for the phonological vs phonetic distinction we have drawn, and show how profound an influence our phonological representations can have on our perception and production of non-native words.

7.6 基于音节的概括

7.6 Syllable-Based Generalizations

我们在 7.1 中说过，音节作为语音成分存在的一些证据来自这样一个事实，即存在重要的语音概括，如果不诉诸音节结构就无法充分表达这些概括。一种这样的概括涉及许多英语口音中软腭化侧音（'dark l's）的分布。对于许多说话者，可能会发现软化 /l/ 和非软化 /l/ 之间的以下分布：

We said in 7.1 that some of the evidence for the existence of the syllable as a phonological constituent comes from the fact that there are significant phonological generalizations which cannot be adequately expressed without appeal to syllable structure. One such generalization concerns the distribution of velarized laterals (‘dark l’s) in many accents of English. For many speakers, the following sort of distribution between velarized and non-velarized /l/ may be found:

(13) 软化和非软化 /l/

gueq07_042

人们可能会尝试按如下方式说明 [l] 和 [ɫ] 的分布：[ɫ] 出现在紧跟另一个辅音时，或出现在单词的末尾（即紧跟单词边界时）。但对这种表述的反对意见之一是，不清楚后续辅音和单词边界可能有什么共同点（如果有的话）。一个更简单的分布陈述，不需要诉诸于这种特殊的环境分离，就是说 [ɫ] 出现在音节的押韵中，而 [l] 出现在首字母. 事实上，我们可能会考虑基于音节的分布来帮助我们诊断lily中第二个 /l/ 的音节状态: 我们可能会说，因为第二个 /l/ 在许多说话者的语音中没有发音，这证实了我们的说法，即它在lily的第二个音节中占据了起始位置，而不是在第一个音节中的尾声位置。

One might attempt to state the distribution of [l] and [ɫ] as follows: [ɫ] occurs when immediately followed by another consonant, or at the end of a word (i.e. when immediately followed by a word boundary). But one of the objections to this formulation is that it is not clear what, if anything, a following consonant and a word boundary might have in common. A simpler statement of the distribution, which does not entail appeal to this peculiar disjunction of environments, is to say that [ɫ] occurs in the rhyme of a syllable, and [l] in the onset. Indeed, we might take that syllable-based account of the distribution to help us diagnose the syllabic status of the second /l/ in lily: we might say that, because the second /l/ is not velarized in the speech of many speakers, this confirms our claim that it occupies onset position in the second syllable of lily, rather than coda position in the first syllable.

让我们考虑另一个例子，来自伦敦英语，一些语音概括的基于音节的性质。一些音韵学家认为元音 [ɒʊ] 和 [ʌʊ] 在许多讲伦敦英语的人的语音中呈互补分布。下表举例说明了这一点：

Let us consider another example, from London English, of the syllable-based nature of some phonological generalizations. The vowels [ɒʊ] and [ʌʊ] are said by some phonologists to be in complementary distribution in the speech of many speakers of London English. The following table exemplifies this:

(14) 伦敦英语中的 [ɒʊ] 和 [ʌʊ]

gueq07_045

声称 [ʌʊ] 音素在其后跟一个 /l/ 时被实现为 [ɒʊ]，它与 [ʌʊ] 在同一音节中（即当后跟一个重复音节的/l/ 时），并且作为 [ ʌʊ] 别处。因此，在左栏的单词中，我们有 [ɒʊ]。在右栏中，load中的元音明显缺少后面的重音节 /l/；至于cola和tombola，我们想根据最大声母的原则来论证，那里的 /l/s 在后面的音节中占据了声母位置，因此那里的 [ʌʊ] 也没有后面的重音节 /l/ .

The claim is that the [ʌʊ] phoneme is realized as [ɒʊ] when it is followed by an /l/ which is in the same syllable as the [ʌʊ] (i.e. when followed by a tautosyllabic /l/), and as [ʌʊ] elsewhere. Thus, in the words in the left-hand column, we have [ɒʊ]. In the right-hand column, the vowel in load clearly lacks a following tautosyllabic /l/; as for cola and tombola, we would want to argue, from the principle of Maximal Onset, that the /l/s there occupy onset position in the following syllable, and thus that the [ʌʊ] there also lacks a following tautosyllabic /l/.

7.7 形态结构、音节结构和重音节化

7.7 Morphological Structure, Syllable Structure and Resyllabification

刚才引用的案例比我们迄今为止所建议的要复杂一些。考虑以下来自相同口音的更多数据：

The case just cited is a little more complex than we have, thus far, suggested. Consider the following further data, from the same accent:

(15)

roller中双元音的音位是什么？一方面，我们已经说过 [ɒʊ] 同位异音出现在重复音节 /l/ 之前。另一方面，Maximal Onset 似乎会让我们将 /l/ 音节化到第二个音节的起始位置。但是如果 /l/ 确实在那个位置，那么我们应该得到 [ʌʊ] 同位异音。那么，为什么我们没有得到那个变位音？让我们考虑两种可能的回应。

What is the phonological status of the diphthong in roller? On the one hand, we have said that the [ɒʊ] allophone appears before a tautosyllabic /l/. On the other hand, it would appear that Maximal Onset would have us syllabify the /l/ into the onset position of the second syllable. But if the /l/ is indeed in that position, then we ought to get the [ʌʊ] allophone. Why, then, do we not get that allophone? Let us consider two possible responses.

一种回应是，说这种口音的人最初有一个简单的语音规则，就像我们已经给出的那样，用于实现 [ʌʊ] 音位，但是随着口音的演变，出现了音位分裂：两个元音本来是互补分布的，但后来出现重叠、平行分布。支持这种观点的证据是最小对的出现，例如holey vs holy。在这里，有人争辩说，我们有明确的证据表明发生了音位分裂。

One response is to say that speakers of this accent originally had a straightforward phonological rule, of the sort we have given, for the realization of the [ʌʊ] phoneme, but that, as the accent has evolved, a phonemic split has emerged: the two vowels were in complementary distribution, but have come to occur in overlapping, parallel distribution. Evidence which is cited in support of this view is the emergence of minimal pairs, such as holey vs holy. Here, it is argued, we have clear evidence that a phonemic split has occurred.

我们可能会对刚才引用的分析提出以下反对意见：它没有注意到关于最小对的成员的重要概括，例如holy和holey，即那些包含 [ʌʊ] 元音的只包含一个语素（它们词素简单），而包含 [ɒʊ] 的词素由多个词素组成（词素复杂）。此外，在每种情况下，相关的元音出现在 /l/ 之前，在形态复杂的情况下，它是词素词尾。

We might make the following objection to the analysis just cited: it is failing to take note of an important generalization concerning the members of minimal pairs such as holy and holey, namely that those containing the [ʌʊ] vowel consist of only one morpheme (they are morphologically simple), whereas those containing [ɒʊ] consist of more than one morpheme (they are morphologically complex). Furthermore, in each case, the relevant vowel occurs before an /l/ which, in the morphologically complex cases, is morpheme-final.

表达语音规则对词法结构的这种敏感性的一种方式是说，在像holey这样的情况下，该规则先于后缀的附加应用，如下所示：

One way of expressing this sensitivity of the phonological rule to morphological structure is to say that the rule applies prior to the affixation of the suffix in cases like holey, as follows:

(16) 词根：/hʌʊl/

(17) [ʌʊ]规则的应用

[ʌʊ] → [ɒʊ] 在重复音节 /l/ 之前

(18) 附加

(19) 再音节化

这种分析需要诉诸进一步的概念：重新音节化。这个想法是，虽然 /l/最初被音节化到尾声位置，但它在附加后被重新音节化，并根据最大起始原则，进入后缀中的空起始（重新音节化到空起始位置的想法作为假设空开始的动机的一部分）。

This analysis requires appeal to a further notion: that of resyllabification. The idea is that, while the /l/ is initially syllabified into coda position, it is resyllabified, after affixation, and according to the Maximal Onset principle, into the empty onset in the suffix (the idea of resyllabification into an empty onset position being part of the motivation for postulating empty onsets).

对于这样的案例，有一个重要的观点需要说明。可能存在语音区分作为最小对（在这种情况下，像holy/holey这样的对）的基础的情况，尽管如此，我们不希望假设两个片段之间存在音位区分。在这种情况下，这些对的成员几乎总是在形态结构上有所不同（在这种情况下，holy在形态上是简单的，而holey在形态上是复杂的），并且这种差异将影响支配音素变位音素的泛化的应用。

There is an important point to be made with respect to cases such as this. It is that there may be cases where a phonetic distinction acts as the basis for minimal pairs (in this case, pairs like holy/holey) where, nonetheless, we do not wish to postulate a phonemic distinction between the two segments. In such cases, the members of those pairs will almost always differ in their morphological structure (in this case, holy is morphologically simple, while holey is morphologically complex), and that difference will affect the application of generalizations which govern the allophones of phonemes.

请注意，在许多非伦敦口音的情况下，语音概括和重音节化的反向排序适用于产生 /i:/ 音素的 [i:] 和 [i:ə] 同位异音的概括情况/l/ 音素的清晰和黑暗的同位异音。我们看到，在许多口音中，在像peel和feel这样的词中，/i:/ 的同位异音是 [i:ə]，而 /l/ 的同位异音是 [ɫ]，如 [fi:əə] ( feel ) . 在像feeling的情况下，/l/ 被重新音节化为后缀 -ɪŋ 的空开头，因此：[fi:lɪŋ]。我们想说的是，支配 /l/ 音素的同位异音的规则（如第 64 页所定义）和支配 /i:/ 音素的同位异音的规则在之后应用重新音节化。

Notice that, in the case of many non-London accents, the reverse ordering of phonological generalizations and resyllabification applies in the case of the generalizations which yield the [i:] and [i:ə] allophones of the /i:/ phoneme and the clear and dark allophones of the /l/ phoneme. We saw that, in many accents, in words like peel and feel, the allophone of /i:/ is [i:ə], and the /l/ allophone is [ɫ], as in [fi:əɫ] (feel). In cases like feeling, the /l/ is resyllabified into the empty onset of the suffix -ɪŋ, thus: [fi:lɪŋ]. We want to say that the rule (as defined on p. 64) which governs the allophones of the /l/ phoneme, and the rule which governs the allophones of the /i:/ phoneme, apply after resyllabification.

我们将认为，虽然音素原则在区分对比和非对比区别时体现了对语音组织本质的重要洞察，但其语音对比的概念有时过于严格，因为它是在没有参考的情况下定义的形态和句法因素对语音组织的影响。因此，我们有时会允许我们的分析可以说是凌驾于音素原则之上。在这样做时，我们并没有放弃在该原则中占有一席之地的概念；相反，我们允许形态因素可能影响语音过程。

We will take the view that, while the phonemic principle, in distinguishing between contrastive and non-contrastive distinctions, embodies an important insight into the nature of phonological organization, its conception of phonological contrast is at times overly restrictive, since it is defined without reference to the influence of morphological and syntactic factors on phonological organization. We will therefore, at times, allow our analyses to, as it were, override the phonemic principle. In doing so, we are not abandoning the notions which find a place in that principle; rather, we are allowing that morphological factors may influence phonological processes.

7.8 总结

7.8 Summing Up

我们在这里采用了将音节作为语音成分的说明。我们已经说过，音节的子部分具有不同程度的感知显着性，因此核心比音节的其他部分更显着。这也许就是为什么在许多语言中，尾辅音经常在狭窄程度上减弱，以至于完全“消失”。这就是许多英语口音中 /ɹ/ 音素的实现所发生的情况，例如：在他们历史的某个阶段， /ɹ/ 是在尾音位置实现的。这也是英语名词历史上许多格结尾（由元音和一个或多个尾辅音组成的后缀）所发生的情况（英语名词曾经有格结尾，如-um、-am、-ans、-uns，所有这些都随着时间的推移而消失了）。

We have adopted here an account of the syllable as a phonological constituent. We have said that the sub-parts of syllables have differing degrees of perceptual salience, so that the nucleus is more salient than the other parts of the syllable. This is perhaps why, in many languages, coda consonants so often diminish in degree of stricture, to the point of ‘fading away’ altogether. This is what has happened to the realizations of the /ɹ/ phoneme in many accents of English, for instance: at one stage in their history, /ɹ/ was realized in coda position. It is also what happened to many case endings (suffixes which consisted of a vowel and one or more coda consonants) in the history of English nouns (English nouns used to have case endings such as -um, -am, -ans, -uns, all of which have disappeared over time).

然而，英语名词中格结尾的丢失涉及另一个因素：不仅核在感知上比尾辅音更显着，而且在英语等语言中，单词中的某些核（强调核）比其他核更显着。我们现在转向英语中单词重音的主题。

The loss of case endings in English nouns involved another factor, however: not only are nuclei more salient perceptually than coda consonants, but, in languages like English, some nuclei in a word (stressed nuclei) are more salient than others. It is to the subject of word stress in English that we now turn.

练习

Exercises

1 对于以下每个单词，说出它是如何音节化的，以及为什么不允许使用替代音节化（例如单词 /kwɒntɪti/（数量）被音化为 /kwɒn.tɪ.ti/；音节化 /kwɒ.ntɪ.ti/违反英语语音规则，因为 /nt/ 不是允许的分支起始；音节化 /kwɒnt.ɪ.ti/ 被最大起始原则排除）：

（一种）	/ɹəvaɪz/	（修改）
（乙）	/pɹədɪkʃən/	（预测）
（C）	/ɹεzɪdεnʃəl/	（住宅）
(四)	/εmpəɹə/	（皇帝）
(五)	/dʒæpəni:z/	（日语）
（F）	/kɒndʌkt/	（行为）

2 在以下单音节语音表示中，说出哪些是英语，哪些是非英语，单音节词的表示。对于每一种非英语形式，说出为什么不可能：

(a) /pnɪt/

(b) /psɪt/

(d) /plεt/

(e) /pɪh/

(f) /ɹpæt/

3 检查以下来自伦敦英语的 Cockney 变体的数据：

（一种）	[lɔɪʔ]	光
（乙）	[laɪdi]	淑女
（C）	[fi:lɪn]	感觉
(四)	[fiɪw]	感觉
(五)	[gɜ:w]	女孩
（F）	[bɒʔu]	瓶子（双音节）
（G）	[filʌ]	填料
（H）	[fiɪwʔʌ]	筛选
（一世）	[waɪ]	方法
(十)	[weɪʔ]	湿

声音 [w] 可能是 /w/ 音素的实现，如 (i) 和 (j) 中。然而，它也可能是 /l/ 音素的同位异音，如 (d)、(e) 和 (h) 中所示。在(f)中，/l/变为音节'[w]'，即[u]。也就是说，/l/ 和 /w/ 之间存在音位重叠。/l/ 的 [w] 变位音出现在什么情况下？你的答案应该用音节结构来表达。

gueqrecard_001

4 进一步的转录练习

在www.wiley.com/go/carrphonetics收听Track 7.1。尽可能详细地转录您在录音中听到的单词，标明音节边界（带句号）和任何音节辅音（例如battle：[bæ.ʔɫ̩]，对于某些说话者）：

狩猎水壶派生的大学破坏假设神圣的宁静也许

8个

8 英语中的节奏和单词重音

Rhythm and Word Stress in English

8.1 英语节奏

8.1 The Rhythm of English

人类说话有节奏：他们通过在语音信号中加入有规律的节拍来进行说话。您可以在诸如The man went to the bar 之类的英语语句中听到这些节拍。在这里，节拍是man，went and bar。在大多数英语变体中，我们不一定要在每个音节上都打一个节拍。在这句话中，没有节拍落在介词to或两次出现的the 上。这是因为与某些其他语言不同，英语是重音节拍：节奏节拍仅落在重音音节上。在我们的例子中，只有人, went和pub都有压力，所以节拍只落在那些人身上。在这方面，英语不同于许多其他语言。以短语Chicken MacNuggets 为例，这是一家知名快餐公司销售的产品的名称。这发音为 [ˌtʃɪkənməkˈnʌɡəts]。此序列中有两个重读音节。（第二个比第一个更突出：我们会回到那个。）序列 [ˈtʃɪkənmək] 和 [ˈnʌɡəts] 在话语中形成节奏单元。这些单位称为公制英尺. 英语中的韵律脚由一个重读音节和后跟零个或多个非重读音节组成。在我们的示例中，第一个韵律脚包含一个重读音节和两个非重读音节：[ˌtʃɪkənmək]。第二韵律脚包含一个重读音节和一个非重读音节：[ˈnʌɡəts]。请注意，韵律英尺之间的分隔不必与单词边界重合：单词边界位于单词Chicken和单词MacNuggets 之间。但节奏边界落在 [ˌtʃɪkənmək] 和 [ˈnʌɡəts] 之间。我们称这些韵律脚为trochaic。这是一个源自名词trochee的形容词。一个 trochee 本质上是一个重音-非重音序列，例如 [ˈnʌɡəts]。我们将检查这些 trochaic第 9 章更详细地介绍了韵律脚。目前，我们只需要注意单词重音模式是英语节奏结构的重要组成部分。

Human beings speak rhythmically: they engage in the act of speaking by putting regular beats in the speech signal. You can hear those beats in an English utterance such as The man went to the bar. Here, the beats are on man, went and bar. In most varieties of English, we do not necessarily place a beat on every single syllable. In this utterance, no beat falls on the preposition to, or on the two occurrences of the. This is because English, unlike certain other languages, is stress-timed: the rhythmic beats fall only on stressed syllables. In our example, only man, went and pub are stressed, so the beats fall only on those. English is unlike many other languages in this respect. Take the phrase Chicken MacNuggets, the name for a product sold by a well-known fast-food company. This is pronounced [ˌtʃɪkənməkˈnʌɡəts]. There are two stressed syllables in this sequence. (The second is more prominent than the first: we’ll come back to that.) The sequences [ˈtʃɪkənmək] and [ˈnʌɡəts] form rhythmic units in the utterance. Those units are called metrical feet. A metrical foot in English consists of a stressed syllable followed by zero or more unstressed syllables. In our example, the first metrical foot contains a stressed syllable and two unstressed syllables: [ˌtʃɪkənmək]. The second metrical foot contains a stressed syllable and one unstressed syllable: [ˈnʌɡəts]. Notice that divisions between the metrical feet need not coincide with word boundaries: the word boundary falls between the words Chicken and the word MacNuggets. But the rhythmic boundary falls between [ˌtʃɪkənmək] and [ˈnʌɡəts]. We call these metrical feet trochaic. This is an adjective derived from the noun trochee. A trochee is essentially a stressed–unstressed sequence, such as [ˈnʌɡəts]. We will examine these trochaic metrical feet in more detail in chapter 9. For the moment, we need only note that word stress patterns are part and parcel of the rhythmic structure of English.

8.2 英文单词重音：完全随机吗？

8.2 English Word Stress: Is It Entirely Random?

我们已经注意到，母语人士的感知能力允许他或她说出一个单词有多少个音节，而无需对音节可能是什么或如何定义有任何有意识的了解。同样，说英语的人可以在不知道“重音”到底是什么的情况下知道单词中哪个音节受到的重音最大。虽然以母语为母语的人可能不知道什么是重音，但很明显，一个音节越重，它在感知上就越突出。例如，大多数以英语为母语的人都会同意，在photography一词中，倒数第二个（倒数第三个）音节（倒数第二个之前的那个，或倒数第二个，一个）是最重读的。同样，大多数演讲者都知道，在袋鼠, 三个音节中的最后一个音节受到的重音最大，依此类推。同样引人注目的是，母语人士可以判断出，虽然袋鼠的最后一个音节比其他任何一个音节都受到更多的压力，但倒数第二个（倒数第三个）音节反过来比倒数第二个（倒数第二个）音节受到更多的压力。photography的最后一个音节和倒数第二个音节是非重读音节，倒数第二个音节之前的音节也是。因此，它们不如具有主要重音的倒数第二个音节那么突出。袋鼠的倒数第二个音节没有重读，是该词中最不显着的音节。让我们假设一个词中受重音最多的音节有主重音，而kangaroo中的倒数第二个音节等音节有副重音；而既没有主要也没有次要重读的音节是非重读音节。因此我们可以说任何给定的词都会有重音模式：在袋鼠的例子中, 一个带有主要重音的最后一个音节，前面是倒数第二个非重音音节，后面是一个带有次要重音的倒数第二个音节。我们可以通过在适当音节的开头放置一个上标变音符号 (ˈ) 来非正式地表示主要重音，并通过使用下标变音符号 (ˌ) 来表示次要重音，让任何非重读音节不带变音符号，如下所示：[ˌkæŋgəˈɹu:] . 该符号用于发音词典。

We have already noted that the native speaker’s perceptual capacities allow him or her to say how many syllables a word has, in the absence of any conscious knowledge of what a syllable might be, or how it might be defined. Similarly, English speakers can tell which syllable in a word receives most stress, in the absence of any conscious knowledge of exactly what ‘stress’ might be. While the native speaker may not know consciously what stress is, it seems clear that, the more stressed a syllable is, the more salient it is, perceptually. For instance, most native speakers of English will agree that, in the word photography, it is the antepenultimate (third last) syllable (the one before the penultimate, or second last, one) which is most stressed. Equally, most speakers will know that, in kangaroo, it is the last of its three syllables which receives most stress, and so on. It is equally striking that the native speaker can judge that, while the final syllable in kangaroo receives more stress than either of the others, the antepenultimate (third from last) syllable in turn receives more stress than the penultimate (second from last) syllable. The final syllable and the penultimate syllable of photography are unstressed, as is the syllable before the ante-penultimate syllable. They are therefore less salient than the antepenultimate syllable, which has primary stress. The penultimate syllable of kangaroo is unstressed and is the least salient syllable in that word. Let us say that the syllable in a word which receives most stress has primary stress, and that syllables such as the ante-penultimate syllable in kangaroo have secondary stress; while syllables which have neither primary nor secondary stress are unstressed syllables. We could therefore say that any given word will have a stress pattern: in the case of kangaroo, a final syllable with primary stress, preceded by a penultimate unstressed syllable, preceded by an antepenultimate syllable with secondary stress. We can informally represent primary stress by placing a superscript diacritic (ˈ) immediately before the start of the appropriate syllable, and secondary stress by using the subscript diacritic (ˌ), leaving any unstressed syllables without a diacritic, as follows: [ˌkæŋgəˈɹu:]. This notation is used in pronouncing dictionaries.

很明显，了解单词的重音模式通常不需要指导，学习母语的孩子通常不会得到父母关于单词重音位置的明确指导。注意到这一点之后，让我们考虑以下问题：说话者如何知道给定单词的重音模式是什么？建议正在学习英语的孩子只需要在学习时记住每个单词的重音模式似乎是合理的。毕竟，对于任何给定的单词，说话者必须记住构成单词音素形式的音素序列。无论是学习英语的儿童还是学习英语的成人，都可以在学习时记住重音模式。有人可能会反对这意味着一个普通的说话者需要记住大量的重音模式（与她或他的词汇表中的单词一样多的重音模式），但我们知道人类非常擅长在大脑中存储大量此类信息记忆。同样，我们可以指出说话者显然必须在心理存储中拥有的大量语音形式；当然，对于具有这种存储能力的有机体来说，存储这些单词的重音模式以及部分构成其语音形式的音素序列并不太难。

It seems clear that knowledge of the stress patterns of words does not normally require instruction, and children acquiring their native language are not normally given explicit instruction by their parents as to where the stresses in a word are placed. Having noted that, let us consider the following question: how does the speaker know what the stress pattern of a given word is? It seems reasonable to suggest that the child who is acquiring English simply has to memorize the stress pattern of each word as it is learned. After all, for any given word, the speaker has to memorize the sequence of phonemes which make up the phonemic form of the word. Whether one is a child acquiring English or an adult learning English, one might just as well memorize the stress pattern while one is at it. One might object that this means that an average speaker has a vast number of stress patterns to memorize (as many stress patterns as there are words in her or his vocabulary), but we know that human beings are very good at storing large amounts of information of this sort in memory. Again, we can point to the vast number of phonological forms which the speaker clearly must have in mental storage; it is, surely, not too tasking to an organism which has that kind of storage capacity to store the stress patterns of those words along with the sequence of phonemes which partly make up its phonological form.

但这并不是说英语中没有无意识地存储的控制重音模式的概括。我们知道，对于某些语言，例如现代希腊语，大多数单词的重音模式完全是任意的。我们还知道有些语言有固定的重音：重音总是落在给定的音节上（例如，在法语中，它总是落在单词的最后一个音节上，而在波兰语中，它总是落在单词的倒数第二个音节上）。让我们考虑一些支持以下观点的证据，即以英语为母语的人无意识地形成了关于英语单词重音模式的概括。拿下面两组双音节英语单词（有两个音节的单词）来说，它们都是形态简单的（即只包含一个语素：没有前缀，没有后缀）：

But this is not to say that there are no unconsciously stored generalizations governing stress patterns in English. We know that, for some languages, such as Modern Greek, the stress pattern of most words is entirely arbitrary. We also know that some languages have fixed stress: the stress always falls on a given syllable (in French, for instance, it always falls on the final syllable of the word, and in Polish on the second last syllable of the word). Let us consider some evidence in favour of the idea that the native speaker of English has unconsciously formed generalizations concerning word stress patterns in English. Take the following two sets of bisyllabic English words (words with two syllables), all of them morphologically simple (i.e. containing only one morpheme: no prefixes, no suffixes):

(1)

第一列的单词重音在倒数第二个音节，第二列的单词重音在最后一个音节。从表面上看，英语中的双音节词似乎主要重音在最后一个或倒数第二个音节上。也许这完全是任意的：也许没有规则。然而，请考虑以下法语单词，它们是由说法语的人和说英语的人说出的，带有明显的法语英语发音（您可以忽略法语转录中任何不熟悉的语音符号：这里重要的是重音模式）：

The words in the first column have primary stress on the penultimate syllable, those in the second column have primary stress on the final syllable. On the face of it, it looks as if bisyllabic words in English may have the primary stress on either the final or the penultimate syllable. Perhaps it’s all entirely arbitrary: perhaps there are no rules. Consider, however, the following French words as uttered by speakers of French and by speakers of English with a noticeably English pronunciation of French (you may ignore any unfamiliar phonetic symbols in the French transcriptions: it’s the stress patterns that matter here):

(2)

法语中的单词重音规则再简单不过了：重读单词的最后一个音节。那么，为什么那么多说英语的人在尝试说法语时，有这种强烈的倾向来重读这么多法语单词中的倒数第二个音节呢？法国的规则并不难理解。而且，毕竟，说英语的人在他们的语言中有无数个双音节词重读在最后一个音节的单词：他们并不是没有习惯将主要重读放在他们自己语言中双音节词的最后一个音节上。那么法语双音节词中这种重音模式的“典型英语”是什么？

The word stress rule in French could not be simpler: stress the final syllable of the word. So why do so many speakers of English have this strong tendency to stress the penultimate syllable in so many French words when they try to speak French? It’s not as if the French rule is hard to grasp. And, after all, English speakers have countless words in their language in which bisyllabic words are stressed on the final syllable: it’s not as if they are not in the habit of placing primary stress on the final syllable of bisyllabic words in their own language. So what is ‘typically English’ about such stress patterns in French bisyllabic words?

还要考虑以下三音节法语单词和典型的英语错误重音：

Consider too the following trisyllabic French words and a typical English mis-stressing of them:

(3)

为什么说英语的人会倾向于在倒数第二个音节上重读这些三音节词？是什么让这种错误的压力成为“典型的英语”？

Why should there be a tendency among English speakers to mis-stress such trisyllabic words on the antepenultimate syllable? What is it that makes such a mis-stressing ‘typically English’?

最后，考虑以下英语中的三音节名词：

Finally, consider the following trisyllabic nouns in English:

(4)

第一列的单词重音在倒数第二个音节，第二列的单词重音在倒数第二个音节，第三列的单词重音在最后一个音节。鉴于单词在句法类别上没有差异（它们都是名词）并且在音节总数上没有差异，看起来好像没有关于英语三音节名词中主词重音的概括：它看起来完全随意，就好像没有规则一样。但是，请考虑以下非英语三音节名词：

The words in the first column have primary stress on the antepenultimate syllable, those in the second column on the penultimate syllable, and those in the third column on the final syllable. Given that the words do not differ as to syntactic category (they are all nouns) and do not differ in terms of total number of syllables, it looks as though there is no generalization concerning primary word stress in English trisyllabic nouns: it all looks entirely arbitrary, as if there were no rules. Consider, however, the following non-English trisyllabic nouns:

(5)

Gigondas moussaka Zaventem tavola

这些单词中的第一个是法语（它是南罗纳河谷的一个城镇和一种葡萄酒的名称），并且像所有法语单词一样重读在最后一个音节上。第二个是希腊语，讲希腊语的人在最后一个音节上强调。第三个是荷兰语（是比利时的一个小镇的名字，也是布鲁塞尔机场的名字）；在荷兰语中，它被强调在倒数第二个音节上。第四个是意大利语中的“桌子”一词，在意大利语中重读倒数第二个音节。

The first of these words is French (it is the name of a town, and a wine, in the Southern Rhône Valley), and is stressed, like all French words, on the final syllable. The second is Greek, and is stressed by Greek speakers on the final syllable. The third is Dutch (it is the name of a town in Belgium, and the name of Brussels airport); it is stressed, in Dutch, on the antepenultimate syllable. The fourth is the Italian word for ‘table’ and is stressed, in Italian, on the antepenultimate syllable.

那些对法语、希腊语、荷兰语或意大利语知之甚少或根本一无所知，并且以前从未听过这些词的英语使用者的惊人之处在于，他们表现出一种非常强烈的倾向，即在第一次遇到这些词时，通过重读这些词来误读这些词倒数第二个音节，如下：[dʒɪˈgɒndəs], [muˈsɒ:kə],[zəˈvεntəm]，[təˈvəʊlə]。如果说英语的人没有单词重音概括，那么这种倾向就非常令人费解，因为这意味着，给定一个以前从未遇到过的单词（尤其是外来词），一个人应该不会表现出偏向于将重音放在任何特定的单词上的倾向音节。人们可能会期望给定的个人在不同场合以三种可能的重音模式中的每一种不同地说出每个词。即使给定的说话者任意地落在给定的压力模式上并此后坚持下去，人们也会期望说话者之间的差异。但这似乎并没有发生：重读第二个音节的发音是他们倾向于选择的发音。你可能会同意，特别是如果你懂法语，荷兰语或意大利语，并且听过说英语的人在这些语言中发音错误，这种发音是“典型的英语”。但是，再一次，什么是“典型的英语”？如果英语中存在单词重音概括，或者至少是单词重音倾向，并且我们知道它们是什么，那么我们只能回答这些问题。我们现在将看看这些概括所采用的形式。

What is striking about English speakers who know little or no French, Greek, Dutch or Italian, and have never heard the words before, is that they show a very strong tendency, on first encountering them, to mispronounce these words by stressing them on the penultimate syllable, as follows: [dʒɪˈgɒndəs], [muˈsɒ:kə], [zəˈvεntəm], [təˈvəʊlə]. If the English speaker has no word stress generalizations, this tendency is deeply puzzling, since that would mean that, given a word one has never encountered before (especially a foreign word), one should display no tendency to prefer placing the stress on any particular syllable. One might expect a given individual to utter each word variably on different occasions, with each of the three possible stress patterns. And even if a given speaker alighted, arbitrarily, on a given stress pattern and stuck to it thereafter, one would expect variation from speaker to speaker. But this does not seem to happen: the pronunciation in which the second syllable is stressed is the one which they tend to opt for. You will probably agree, especially if you know any French, Dutch or Italian and have heard English speakers mispronouncing words in those languages, that this kind of pronunciation is ‘typically English’. But, once again, what is ‘typically English’ about it? We can only answer such questions if there are word stress generalizations, or at least word stress tendencies, in English, and if we know what they are. We will now look at the form those generalizations take.

8.3 英语单词重音：一些一般原则

8.3 English Word Stress: Some General Principles

第一个一般原则（原则 1：基于结尾的原则）是英语单词中主重音的位置是通过从词尾开始计算的：单词中的主重音倾向于落在最后一个音节上单词的倒数第二个（倒数第二个）音节或倒数第二个（倒数第三个）音节（尽管它可以比那个更早）。这反映了这样一个事实，即大多数英语变体的单词重音模式本质上都是trochaic。回想一下，形容词trochaic源自名词trochee, 并且 trochee 是一个重读音节（无论是主重读音节还是次要重读音节）后跟零个或多个非重读音节。我们会说重读的单音节词（如box）、倒数第二重音的词（如spider和departure）和倒数第二重音的词（如cinema和America）都表现出次序重音模式。大多数英语变体的节奏都是三段式的：有一种趋势是将节奏节拍放在三段式脚的重读音节上。

A first general principle (Principle 1: The End-Based Principle) is that the placement of primary stresses in English words is calculated by counting from the end of the word: the primary stress in a word will tend to fall on either the final syllable of the word, the penultimate (second last) syllable or the antepenultimate (third last) syllable (though it can fall earlier than that). This reflects the fact that most varieties of English have word stress patterns which are essentially trochaic. Recall that the adjective trochaic is derived from the noun trochee, and that a trochee is a stressed syllable (whether primary stressed or secondary stressed) followed by zero or more unstressed syllables. We will say that stressed monosyllabic words (such as box), words with penultimate stress (such as spider and departure) and words with antepenultimate stress (such as cinema and America) all exhibit trochaic stress patterns. The rhythm of most varieties of English is trochaic: there is a tendency to place the rhythmic beat on the stressed syllables of trochaic feet.

第二个一般原则（原则 2：节奏原则）是，虽然英语单词可以以多达四个非重读音节结尾（如un ˈ gentlemanliness ），但英语单词不能以一个以上的非重读音节开头。原则 2 与原则 1 直接相关：我们不以两个或多个非重读音节的序列开始单词的原因是，如果我们在这些序列中放置次要重音，我们可以创建一个 trochaic foot，这是从英语韵律结构的观点。当我们从Japan派生日语时，主要重音从最后一个音节转移日本到日语的最后一个音节（我们将很快研究这种重音转移）。但是，在转移了主要重音之后，我们不能在主要重音音节之前留下两个非重音音节序列的日语单词：我们必须将次要重音放在前面两个音节之一上：ˌ Japa ˈnese 。

A second general principle (Principle 2: The Rhythmic Principle) is that, while it is possible for English words to end with as many as four unstressed syllables (as in unˈgentlemanliness), English words cannot begin with more than one unstressed syllable. Principle 2 is directly related to Principle 1: the reason why we do not begin words with sequences of two or more unstressed syllables is that, if we place a secondary stress within such sequences, we can create a trochaic foot, which is desirable from the point of view of the rhythmic structure of English. When we derive Japanese from Japan, the primary stress shifts from the final syllable of Japan onto the final syllable of Japanese (we will examine such stress shifts shortly). But, having shifted the primary stress, we cannot leave the word Japanese with a sequence of two unstressed syllables preceding the primary stressed syllable: we must place a secondary stress on one of the two preceding syllables: ˌJapaˈnese.

当这种情况发生时，第三个一般原则（原则 3：派生原则）开始发挥作用：有一种趋势是将次要重音放在派生词中具有主要重音的音节上（我们正在派生更复杂的词）字从）。例如，characterization一词从其派生词characterize（本身源自character）表现出主重音的转变。由于原则 2 规定我们不能离开表征这个词在主重音音节之前有一系列四个非重音音节，我们必须在该音节序列中的某个地方放置一个次要重音。派生原则说我们应该把它放在派生词中有主重音的音节上：ˌ characteri ˈ zation 。

When this happens, a third general principle (Principle 3: The Derivational Principle) comes into play: there is a tendency to place the secondary stress on the syllable which had primary stress in the deriving word (the word which we are deriving the more complex word from). For instance, the word characterization exhibits a shift of primary stress from its deriving word ˈcharacterize (which itself is derived from ˈcharacter). Since Principle 2 dictates that we cannot leave the word characterization with a series of four unstressed syllables prior to the syllable with primary stress, we must place a secondary stress somewhere in that sequence of syllables. The Derivational Principle says that we ought to place it on the syllable which had primary stress in the deriving word: ˌcharacteriˈzation.

然而，原则 3 可能会被第四条一般原则（原则 4：重音冲突避免原则）所否决，该原则指出应该尽量避免有两个相邻的重音音节。在 ˌ characteriation zation的情况下，这两个原则都得到遵守。但请注意，在 ˌ Japa ˈnese的情况下，情况并非如此：派生词Ja ˈ pan的主要重音落在最后一个音节上。原则 3 规定我们将次重音放在该音节上，但原则 4 规定我们不这样做，因为这会导致重音冲突，两个相邻的重音音节：Jaˌpaˈnese. 像这样的案例表明，当派生原则和重音冲突避免原则发生冲突时，我们可以选择将次要重音放在哪个音节上，但重音冲突避免原则 4 占主导地位：简而言之就像 ˌ Japa ˈ nese一样，我们将次重音放在倒数第二个音节上，以满足重音冲突避免约束。避免重音冲突原则是英语中一个强烈的普遍倾向。但是英语中有些词违反了这个原则，比如动词ˌreˈrun，名词ˌchamˈpagne和ˌDunˈ dee . 我们将在下一章中更深入地研究这些。就派生原则而言，有Burmese、Chinese等词，是由“ Burma和“ China ”派生出来的。在这种情况下，我们别无选择，只能将次重音放在倒数第二个音节上：ˌ Bur ˈ mese , ˌ Chi ˈ nese 。但是我们可以在两个或多个音节之间进行选择，例如 ˌ Japa ˈnese和ˌ Portu ˈguese，我们既避免了重音冲突，又在单词中创建了两个 trochaic 脚的序列。因此，原则 4，重音冲突避免原则，通常与原则 2，节奏原则一起使用，在单词的开头创建额外的 trochaic 脚。为什么尽管避免了单词开头的轻音节序列，但在单词结尾处仍然可以有多达四个轻音节，就像绅士风度一样？因为它们以某种后缀结尾。我们很快就会考虑这样的后缀。现在，让我们进一步看看形态简单的单词中的单词重音。

However, Principle 3 may be overruled by a fourth general principle (Principle 4: The Stress Clash Avoidance Principle), which states that one should try to avoid having two adjacent stressed syllables. In the case of ˌcharacteriˈzation, both principles are adhered to. But note that, in the case of ˌJapaˈnese, this is not the case: the primary stress in the deriving word Jaˈpan falls on the final syllable. Principle 3 dictates that we place the secondary stress on that syllable, but Principle 4 dictates that we do not, since this would result in a stress clash, with two adjacent stressed syllables: Jaˌpaˈnese. Cases like this demonstrate that, where the Derivational Principle and the Stress Clash Avoidance Principle come into conflict, and we have an option as to which syllable to place the secondary stress on, it is the Stress Clash Avoidance Principle 4 which predominates: in a word like ˌJapaˈnese, we place the secondary stress on the antepenultimate syllable in order to satisfy the Stress Clash Avoidance constraint. The Stress Clash Avoidance Principle is a strong general tendency in English. But there are words in English which violate that principle, such as the verb ˌreˈrun, and the nouns ˌchamˈpagne and ˌDunˈdee. We will look at these in more depth in the following chapter. As far as the Derivational Principle is concerned, there are words such as Burmese and Chinese, which are derived from ˈBurma and ˈChina. In such cases, we have no choice but to place the secondary stress on the penultimate syllable: ˌBurˈmese, ˌChiˈnese. But where we have a choice between two or more syllables, as in ˌJapaˈnese and ˌPortuˈguese, we both avoid a stress clash and also create a sequence of two trochaic feet within the word. So Principle 4, the Stress Clash Avoidance Principle, often works together with Principle 2, the Rhythmic Principle, to create extra trochaic feet at the beginnings of words. Why, despite the avoidance of sequences of unstressed syllables at the beginning of a word, is it nonetheless possible to have as many as four unstressed syllables at the end of a word, as in gentlemanliness? Because they end with certain kinds of suffix. We will consider such suffixes shortly. For the moment, let us look further at words stress in morphologically simple words.

8.4 词法简单词的词重音分配

8.4 Word Stress Assignment in Morphologically Simple Words

英语是日耳曼语系的语言，它从拉丁语言中借用了大量词汇，尤其是法语和拉丁语，其中许多词汇都带有拉丁语后缀和前缀。这样做的结果是使单词的重音模式比原本更复杂，非母语人士将证明他们在试图掌握英语单词的重音模式时经常遇到的困难。尽管如此，英语单词重音模式还是有相当大的规律性。让我们首先考虑现代英语中明显没有前缀或后缀的单词（尽管其中一些在历史上有前缀）。我们将区分词汇类别的词和非词汇类别的词。词汇类别的词是名词、动词、形容词和副词。非词汇类别的词包括介词、限定词（如the、this、his）、代词（如he、her) 和连词and。非词汇类别的词，通常称为功能词，通常不重读。在词汇类别的单词中，主要重音位置可能会有所不同，具体取决于单词的句法类别。

English is a Germanic language which has borrowed a huge amount of vocabulary from Latinate languages, notably French and Latin, many of them with Latinate suffixes and prefixes. The effect of this has been to make the word stress patterns more complex than they would otherwise have been, and non-native speakers will testify to the difficulty they often experience in trying to master the stress patterns of English words. Nonetheless, there is considerable regularity in English word stress patterns. Let us begin by considering words which clearly do not have prefixes or suffixes, in present-day English (though some of these had prefixes historically). We will distinguish words of a lexical category from words of a non-lexical category. Words of a lexical category are nouns, verbs, adjectives and adverbs. Words of a non-lexical category include prepositions, determiners (such as the, this, his), pronouns (such as he, her) and the conjunction and. Words of a non-lexical category, often referred to as function words, are not normally stressed. Among the words of a lexical category, primary stress placement may vary, depending on the syntactic category of the word.

词汇类别（例如box、run、big）的单音节词是没有问题的：只有一个音节落在主重音上。因此，让我们继续研究形态简单的双音节词，然后继续研究形态简单的多音节词（具有三个或更多音节的词）。

Monosyllabic words of a lexical category (such as box, run, big), are unproblematic: there is only one syllable for the primary stress to fall on. Let us therefore move on to morphologically simple bisyllabic words, and then proceed to morphologically simple polysyllabic words (words with three or more syllables).

8.4.1 形态简单的双音节词

8.4.1 Morphologically Simple Bisyllabic Words

8.4.1.1 双音节名词

8.4.1.1 Bisyllabic Nouns

这里的基本模式是本土日耳曼语的托洛伊格模式，即主要重音在倒数第二个音节上，如Arab、bigot、carpet、district、effort、female、gremlin、harbour、lemon、market、native、person、rabbit、参议院，萝卜，吸血鬼，女人。然而，这种基本模式有很多例外，其中双音节名词有最后的重音。其中有从其他语言借用并用双元音字母书写的名词，如balloon、bazaar、canteen、harpoon、marquee、papoose、raccoon、shampoo、taboo、veneer。另一组强调最后一个音节的双音节名词是从法语借来的；它们包含法语结尾 -ee , - ette , - ade , - elle , - esse , - asse , - eur , - euse，如grandee、gazette 、 parade 、 gazelle 、 finesse 、 crevasse 、 liqueur 、 masseuse 。由于法语中的单词重音总是落在最后一个音节上，因此诸如此类的单词大多保留了法语的重音模式。其他保留了词尾重音的法语双音节外来词包括酒店：[həʊˈtʰεɫ]。如果这个词被完全本土化，它会发音为 [ˈhəʊtəə]，带有日耳曼语的原生托查音重音模式。我们在这里假设诸如gazette、parade之类的词等在现代英语中大多形态简单。^1个

The basic pattern here is the native Germanic trochaic pattern, that is, with primary stress on the penultimate syllable, as in Arab, bigot, carpet, district, effort, female, gremlin, harbour, lemon, market, native, person, rabbit, senate, turnip, vampire, woman. However, there is a substantial class of exceptions to this basic pattern, in which bisyllabic nouns have final stress. Among these are nouns which have been borrowed from other languages and are written with a double vowel letter, as in balloon, bazaar, canteen, harpoon, marquee, papoose, raccoon, shampoo, taboo, veneer. Another set of bisyllabic nouns stressed on the final syllable have been borrowed from French; they contain the French endings -ee, -ette, -ade, -elle, -esse, -asse, -eur, -euse, as in grandee, gazette, parade, gazelle, finesse, crevasse, liqueur, masseuse. Since word stress in French always falls on the final syllable, words such as these have mostly retained the French stress pattern. Other French bisyllabic loanwords which have retained final stress include hotel: [həʊˈtʰεɫ]. Had this word been fully nativized, it would be pronounced [ˈhəʊtəɫ], with the native Germanic trochaic stress pattern. We assume here that words such as gazette, parade, etc. are mostly morphologically simple in contemporary English.1

8.4.1.2 双音节形容词

8.4.1.2 Bisyllabic Adjectives

日耳曼语的基本模式再次是 trochaic，即重音在倒数第二个，如angry, brilliant, central, crazy, dozy, fragile, frigid, happy, honest, lazy, modest, narrow, orange, purple, sudden, timid, urgent ，黄色。在以 - ic结尾的形态简单的双音节形容词中也可以找到 trochaic 模式，例如cosmic，疯狂的，北欧的，静态的。²然而，也有带词尾重音的双音节形容词，如complete、implement、intense、precise、select。这些包含起源于拉丁语的历史前缀：它们来自拉丁语或法语。这些历史前缀中的大多数在当代英语中不再算作生产性前缀（它们不能与词根自由组合形成新词）。对于绝大多数说英语的人来说，这些词中的大量现在在形态上都很简单。除此之外，历史上包含法语后缀的双音节形容词，例如bizarre和grotesque，都有词尾重音。与gazette和gazette等名词一样parade，我们在这里假设bizarre和grotesque等形容词在形态上很简单。

The basic Germanic pattern is again trochaic, i.e. with stress on the penult, as in angry, brilliant, central, crazy, dozy, fragile, frigid, happy, honest, lazy, modest, narrow, orange, purple, sudden, timid, urgent, yellow. The trochaic pattern can also be found in morphologically simple bisyllabic adjectives ending in -ic, as in cosmic, frantic, Nordic, static.2 However, there are bisyllabic adjectives with final stress, such as complete, immense, intense, precise, select. These contain historical prefixes which are Latinate in origin: they come from Latin or French. Most of these historical prefixes no longer count as productive prefixes in contemporary English (they cannot be freely combined with roots to form new words). Large numbers of these words are now morphologically simple, for the vast majority of speakers of English speakers. In addition to these, bisyllabic adjectives which contain what were, historically, French suffixes, such as bizarre and grotesque, have final stress. As with nouns like gazette and parade, we assume here that adjectives such as bizarre and grotesque are morphologically simple.

8.4.1.3 双音节副词

8.4.1.3 Bisyllabic Adverbs

再一次，基本的日耳曼模式是 trochaic。许多双音节副词以 in- ly结尾，如slowly和quickly。我们将在 8.5 中处理这些问题。其他不以 in- ly结尾的副词包括rather和very。这些也有一个 trochaic 应力模式。

Once again, the basic Germanic pattern is trochaic. Many bisyllabic adverbs end in -ly, as in slowly and quickly. We will deal with those in 8.5. Other adverbs which do not end in -ly include rather and very. These also have a trochaic stress pattern.

8.4.1.4 双音节动词

8.4.1.4 Bisyllabic Verbs

在大多数英语变体中，双音节动词的基本三重音倾向在双音节动词中不太明显：如我们所见，有许多词尾有主重音：compact、detract、deny、export、impose、object、permit、produce、record、subsume、违背。与形容词一样，其中许多在历史上都有借用它们的语言的前缀：法语或拉丁语（例如compact、de-tract、ex-port、im-pose、ob-ject、re-cord、sub-sume ). 避免强调历史前缀的趋势很强烈：令人惊讶的是，此类双音节动词与其双音节名词对应物经常不同，如动词-名词对con ˈ tract (verb) but ˈ contract (noun), diˈ gest (动词) but ˈ digest (名词), pro ˈ duce (动词) but ˈ pro duce (名词), ex ˈ port (动词) but ˈ export (名词), com ˈ pound (动词) but ˈ compound (名词) ), 等等，其中动词采用最后重音，但名词采用正常的倒数第二重音。（但这种模式也有例外，如在con ˈ cern中，名词和动词都采用动词模式，而ˈ preface中，名词和动词都采用名词模式。）

The basic trochaic tendency in most varieties of English is much less evident in bisyllabic verbs: there are many with final primary stress, as we have seen: compact, detract, deny, export, impose, object, permit, produce, record, subsume, transgress. As with adjectives, many of these historically had prefixes in the languages they were borrowed from: French or Latin (e.g. com-pact, de-tract, ex-port, im-pose, ob-ject, re-cord, sub-sume). The tendency to avoid stressing the historical prefix is strong: it is striking that such bisyllabic verbs differ so often from their bisyllabic noun counterparts, as in the verb–noun pairs conˈtract (verb) but ˈcontract (noun), diˈgest (verb) but ˈdigest (noun), proˈduce (verb) but ˈproduce (noun), exˈport (verb) but ˈexport (noun), comˈpound (verb) but ˈcompound (noun), etc., where the verb takes final stress but the noun takes the normal trochaic penultimate stress. (But there are exceptions to this pattern, as in conˈcern where the noun and verb both take the verb pattern, and ˈpreface, where the noun and verb both take the noun pattern.)

还值得注意的是，以-ate 结尾的双音节动词将主要重读-ate，如create、deflate、locate、migrate、placate、sedate。这些必须与三个或三个以上音节的词区分开来（见下文）。请注意，在 GA 中，其中一些词采用倒数第二重音，例如frustrate和locate。

It is also worth noting that bisyllabic verbs ending in -ate will take primary stress on the -ate, as in create, deflate, locate, migrate, placate, sedate. These have to be distinguished from -ate words with three or more syllables (see below). Note that, in GA, some of these words take penultimate stress, such as frustrate and locate.

然而，有些双音节动词具有基本的三重音模式，例如argue、canter、dither、enter、equal、falter、gather、govern、hurry、manage、market、marry、narrow、rattle、sully、travel。

There are, however, bisyllabic verbs with the basic trochaic stress pattern, such as argue, canter, dither, enter, equal, falter, gather, govern, hurry, manage, market, marry, narrow, rattle, sully, travel.

8.4.2 形态简单的多音节词

8.4.2 Morphologically Simple Polysyllabic Words

8.4.2.1 多音节名词

8.4.2.1 Polysyllabic Nouns

基本的日耳曼模式是 trochaic。对于多于两个音节的单词，这意味着重读倒数第二个音节，如在academy 中，美国、羚羊、相机、电影院、保管、赤字、大象、皇帝、和谐、图书馆、旋律、天堂、数量、策略。然而，这个基本模式有很多例外，其中多音节名词有词尾重音。其中有名词的最后一个音节用双元音字母书写，如kangaroo。另一组强调最后一个音节的多音节名词是从法语借来的；它们包含法语结尾 - ette、-ade、-elle、-esque、-eur，如cigarette、lemonade、bagatelle、picturesque、connoisseur。诸如此类的词在现代英语中大多形态简单。也就是说，它们并没有真正的后缀：while there is a morpheme cigar，例如，香烟不是小雪茄。^3个

The basic Germanic pattern is trochaic. For words of more than two syllables, this means having primary stress on the antepenultimate syllable, as in academy, America, antelope, camera, cinema, custody, deficit, elephant, emperor, harmony, library, melody, paradise, quantity, strategy. However, there is a substantial class of exceptions to this basic pattern, in which polysyllabic nouns have final stress. Among these are nouns which have final syllables which are written with a double vowel letter, as in kangaroo. Another set of polysyllabic nouns stressed on the final syllable have been borrowed from French; they contain the French endings -ette, -ade, -elle, -esque, -eur, as in cigarette, lemonade, bagatelle, picturesque, connoisseur. Words such as these are mostly morphologically simple in contemporary English. That is, they do not really have a suffix: while there is a morpheme cigar, a cigarette is not a small cigar, for instance.3

有一组名词在倒数第二个元音后有辅音丛，这些名词有倒数第二个主重音，如advantage、apartment、consensus、disaster、objective。还有一组名词具有三个或更多音节并且以-ics结尾。这些也往往具有倒数第二的压力，如声学、电学、语言学、物流、数学、统计学。

There is a set of nouns which have consonant clusters after the penultimate vowel, and these have penultimate primary stress, as in advantage, apartment, consensus, disaster, objective. There is also a set of nouns which have three or more syllables and which end in -ics. These too tend to have penultimate stress, as in acoustics, electrics, linguistics, logistics, mathematics, statistics.

最后，还有一组以元音结尾的外来词，与基本的倒数第二个模式不同，倒数第二个重音，例如banana、bikini、chapatti、chorizo、karate、martini、moussaka、mosquito、potato、samosa、tomato、volcano .

Finally, there is a set of loanwords ending in a vowel which depart from the basic antepenultimate pattern, and take penultimate stress, such as banana, bikini, chapatti, chorizo, karate, martini, moussaka, mosquito, potato, samosa, tomato, volcano.

8.4.2.2 多音节形容词

8.4.2.2 Polysyllabic Adjectives

同样，基本模式是trochaic，具有倒数第二重音，如一般情况下，聪明，幼稚，沉默寡言。然而，有一组形容词在倒数第二个元音后有一个辅音丛，这些形容词倒数第二个重音，如独立的、灾难性的、放纵的、秘密的、重要的、客观的、巨大的。其中包括在 rhotic 口音中具有“rC”簇的单词，以及曾经在非 rhotic 口音中具有“rC”簇的单词，例如巨大的、母性的。

Again, the basic pattern is trochaic, with antepenultimate stress, as in general, intelligent, juvenile, taciturn. However, there is a set of adjectives with a consonant cluster after the penultimate vowel, and these take penultimate stress, as in dependent, disastrous, indulgent, clandestine, momentous, objective, tremendous. These include words which have an ‘rC’ cluster in rhotic accents, and which used to have an ‘rC’ cluster in non-rhotic accents, as in enormous, maternal.

以-ate结尾的多音节形容词有倒数第二个主重音，如deliberate、elaborate、lucky、indicate、legitimate。因此，这些不同于以 - ate结尾的双音节动词，后者的主重音放在 - ate上（但美国说话者在其中一些词中有倒数第二个主重音，正如我们所见）。

Polysyllabic adjectives ending in -ate have antepenultimate primary stress, as in deliberate, elaborate, fortunate, inadequate, legitimate. These are therefore unlike bisyllabic verbs ending in -ate, which have primary stress on the -ate (but American speakers have penultimate primary stress in some of these words, as we have seen).

8.4.2.3 多音节动词

8.4.2.3 Polysyllabic Verbs

与双音节动词一样，多音节动词经常蔑视基本的三音节英语单词重音模式：有许多具有三个或更多音节的动词具有最后的主要重音，如entertainment、intervene、intersect。与双音节动词一样，其中许多动词都有起源于拉丁语的词源前缀（例如enter - 和inter-）。

As with bisyllabic verbs, polysyllabic verbs often flout the basic trochaic English word stress pattern: there are many verbs with three or more syllables which have final primary stress, as in entertain, intervene, intersect. As with bisyllabic verbs, many of these have etymological prefixes which are Latinate in origin (such as enter- and inter-).

与以 - ate 结尾的双音节动词不同，以 - ate结尾的多音节动词遵循基本的倒数第二个主要重音模式，如coordinate , deliberate, elaborate, investigate, originate。因此，这些动词与以 in- ate结尾的多音节形容词具有相同的重音模式。

Unlike bisyllabic verbs ending in -ate, polysyllabic verbs ending in -ate follow the basic antepenultimate primary stress pattern, as co-ordinate, deliberate, elaborate, investigate, originate. These verbs thus have the same stress pattern as polysyllabic adjectives ending in -ate.

我们现在已经确定了四个因素，这些因素可能在英语中形态简单的单词的单词重音分配中发挥作用。首先，句法范畴这个词可能会发挥作用：例如，我们看到许多动词偏离了基本的 trochaic 模式。其次，我们看到拉丁语起源的历史（词源）前缀的存在会影响单词的重音模式：我们看到拉丁语词缀如ex-、pro-、inter-，尽管它们大多不再是生产性前缀在现代英语中，通常不带主重音。第三，我们看到拼写在单词重音中起作用。例如，shampoo、papoose和kangaroo等词都在包含双元音字母的最后一个音节上表现出主要重音。与此相关的是第四个因素，即英语中外来词的存在：诸如shampoo和hotel之类的词保留了他们从中借用的语言的重音模式。另请注意，外来词和拉丁词缀的存在都在 gazelle 、grotesque、cigarette、picturesque和connoisseur等词的重音模式中起作用。

We have now identified four factors which may play a part in word stress assignment in morphologically simple words in English. Firstly, the syntactic category of the word may play a role: we saw, for instance, that many verbs depart from the basic trochaic pattern. Secondly, we saw that the presence of historical (etymological) prefixes of a Latinate origin can affect the stress pattern of a word: we saw that Latinate affixes such as ex-, pro-, inter-, although they are mostly no longer productive prefixes in present-day English, typically fail to take primary stress. Thirdly, we saw that spelling plays a role in word stress. For instance, words such as shampoo, papoose and kangaroo all exhibit primary stress on a final syllable containing a double vowel letter. Connected with this is a fourth factor, i.e. the presence of loanwords in English: words such a shampoo and hotel have retained the stress pattern of the language they were borrowed from. Note too that both loanwords and the existence of Latinate affixes are at work in the stress patterns of words such as gazelle, grotesque, cigarette, picturesque and connoisseur.

我们现在可以回到我们在 8.2 中考虑的压力模式。回想一下，我们注意到说英语的人倾向于通过将重音放在倒数第二个音节上来错误重读双音节法语单词，例如manger（“吃”）和bâteau （“船”）。我们现在可以明白其中的原因了：这是因为这是英语中双音节词的基本模式。我们还注意到，说英语的人倾向于将主词重音放在多音节法语单词的倒数第二个音节上，例如fermeture（'closure'）和bâtiment（'building'）。原因是这是英语多音节词的基本模式。这就是典型的关于那些错误强调的英语。在多音节词的发音中，例如在Gigondas和Zaventem中，拼写起着一定的作用：英语单词倒数第二个音节的拼写中存在两个辅音字母，通常会将重音从基本的倒数第二重音模式转移到倒数第二重音模式。最后，诸如moussaka和tavola之类的多音节词属于以元音结尾的外来词类别，元音特别是倒数第二个主重音。

We may now return to the stress patterns we considered in 8.2. Recall that we noted that English speakers tend to mis-stress bisyllabic French words such as manger (‘to eat’) and bâteau (‘boat’) by placing the stress on the penultimate syllable. We can now see the reason for this: it is because that is the basic pattern for bisyllabic words in English. We also noted that English speakers tend to place primary words stress on the antepenultimate syllable of polysyllabic French words such as fermeture (‘closure’) and bâtiment (‘building’). The reason is that this is the basic pattern for English polysyllabic words. That is what is typically English about those mis-stressings. In the pronunciation of polysyllabic words such as Gigondas and Zaventem, the spelling plays a role: the existence of two consonant letters in the spelling of the penultimate syllable of the word in English often shifts the stress away from the basic antepenultimate stress pattern to a penultimate stress pattern. Finally, polysyllabic words such as moussaka and tavola fall within the class of loanwords ending in a vowel which exceptionally take penultimate primary stress.

现在让我们考虑形态复杂的单词中的单词重音。

Let us now consider word stress in morphologically complex words.

8.5 单词重音分配和形态结构

8.5 Word Stress Assignment and Morphological Structure

英语既有后缀（如在sorted中，带有-ed后缀）也有前缀（如在indirect中，带有前缀in-）。让我们从后缀开始。这些可以细分为屈折后缀和派生后缀。如果没有添加后缀，通常会说添加屈折后缀会产生“不同形式”的单词。例如，当后缀-ing被添加到动词obscure时，结果词obscuring（如He's obscuring the issue）被认为是一种形式那个动词；当复数后缀添加到名词tractor时，生成的单词tractors是该名词的一种形式。但是当派生后缀被添加到一个词时，据说产生的不是同一个词的不同形式，而是另一个词。因此，当后缀-ly被添加到形容词时，比如bold，结果副词boldly是一个不同的词。同样，当形容词加上后缀-ness时，如boldness，结果是一个不同的词。英语中派生后缀的其他例子是 - ity（如个人/个性），- ee（如离婚/离婚），- al（如个人/个人），- ian（如在Wagner/Wagnerian中），- ic（如在atom/atomic中），- ish（如在green/greenish中），- y（如在sleep/sleepy中）等。屈折后缀不重读，并且有对英语单词中的单词重音没有影响，从诸如ˈ refuge / ˈ refuges（复数后缀）， ˈ comment / ˈ commented（过去式后缀），de ˈ velop / ˈ developing , （进行式后缀），和ˈ varnish / ˈ varnishes（现在时后缀）。

English has both suffixes (as in sorted, with the -ed suffix) and prefixes (as in indirect, with the prefix in-). Let us begin with suffixes. These may be subdivided into inflectional and derivational suffixes. The addition of an inflectional suffix is often said to produce ‘a different form’ of the word one would have if the suffix had not been added. For instance, when the suffix -ing is added to the verb obscure, the resulting word, obscuring (as in He’s obscuring the issue), is said to be a form of that verb; when the plural suffix is added to the noun tractor, the resulting word tractors is a form of that noun. But when a derivational suffix is added to a word, it is said to produce not a different form of the same word, but another word. Thus, when the suffix -ly is added to an adjective, say bold, the result, the adverb boldly, is a distinct word. Similarly, when the suffix -ness is added to an adjective, as in boldness, the result is a distinct word. Other examples of derivational suffixes in English are -ity (as in personal/personality), -ee (as in divorce/divorcee), -al (as in person/personal), -ian (as in Wagner/Wagnerian), -ic (as in atom/atomic), -ish (as in green/greenish), -y (as in sleep/sleepy), etc. Inflectional suffixes are not stressed, and have no effect on word stress in English words, as can be seen from pairs such as ˈrefuge/ˈrefuges (plural suffix), ˈcomment/ˈcommented (past tense suffix), deˈvelop/ˈdeveloping, (progressive suffix), and ˈvarnish/ˈvarnishes (present tense suffix).

在英语派生后缀中，有的加在单词上时对重音没有影响，有的则影响重音模式。这两类后缀分别称为应力中性和应力转移后缀。重音转移后缀都是拉丁血统，即来自法语或拉丁语。但并非所有拉丁语后缀都是重音转移。让我们从考虑重音中性后缀开始。

Among the English derivational suffixes, some have no effect on stress when added to a word, while others do affect the stress pattern. These two classes of suffix are referred to as the stress-neutral and stress-shifting suffixes, respectively. The stress-shifting suffixes are all of Latinate descent, i.e. from French or Latin. But not all Latinate suffixes are stress-shifting. Let us begin by considering stress-neutral suffixes.

8.5.1 重音中性后缀

8.5.1 Stress-Neutral Suffixes

英语是日耳曼语；所有本土日耳曼语后缀都是重音中性的。这些包括副词后缀 - ly，如ˈ brightly , ˈ deeply , ˈ dimly , ˈ madly , ˈ quickly , ˈ slowly , ˈ truly等。我们可以看到，添加后缀后重音没有变化派生副词的单音节形容词（（ bright , deep , dim等）。因此，这些副词的重音模式是英语双音节词的基本双音节模式：重音在倒数第二个音节上。多音节副词以 - ly结尾包括cheerily、 happily、in credibly 、re markably 、tre mendously 、 wearily 。同样，主要重音仍然保留在副词派生的形容词中： ˈ cheery , ˈ happy , in ˈ credible , re ˈ markable , tre ˈ mendous , ˈ weary. 在所有这些情况下，所产生的副词都是一个多音节词，具有多音节词的基本三节段模式，即倒数第二重音。但是，如果派生形容词的主重音位于其倒数第二个音节或更早的位置，则主重音将比倒数第二个音节下降得更早，例如ˈ comfortable，派生自ˈ comfortable，它在 RP 中具有三个音节，但在 GA 中具有四个音节。请注意，诸如comfortable和idle之类的形容词具有音节 /l/，如 [ˈaɪdɪ̩]，在添加-ly时将失去一个音节：idly发音为 [ˈaɪdli]。

English is a Germanic language; all native Germanic suffixes are stress-neutral. These include the adverbial suffix -ly, as in the words ˈbrightly, ˈdeeply, ˈdimly, ˈmadly, ˈquickly, ˈslowly, ˈtruly, etc. We can see here that the stress does not shift when the suffix is added to the monosyllabic adjective from which the adverb is derived ((ˈbright, ˈdeep, ˈdim, etc.). The stress pattern of these adverbs is therefore the basic trochaic pattern for English bisyllabic words: stress on the penultimate syllable. Polysyllabic adverbs ending in -ly include ˈcheerily, ˈhappily, inˈcredibly, reˈmarkably, treˈmendously, ˈwearily. Again, the primary stress remains where it was in the adjective which the adverb is derived from: ˈcheery, ˈhappy, inˈcredible, reˈmarkable, treˈmendous, ˈweary. In all of these cases, the resulting adverb is a polysyllabic word with the basic trochaic pattern for polysyllabic words, i.e. antepenultimate stress. However, the primary stress will fall earlier than the antepenultimate syllable if the deriving adjective has primary stress on its antepenultimate syllable or earlier, as in ˈcomfortably, derived from ˈcomfortable which has three syllables in RP, but four in GA. Note that adjectives such as comfortable and idle, which have a syllabic /l/, as in [ˈaɪdɫ̩], will lose a syllable when -ly is added: idly is pronounced [ˈaɪdli].

本土日耳曼语形容词后缀，如-er、-est、-ish、-ful、-less、-y也对重音没有影响，因此当它们被添加到单音节词根时，结果是一个长音重音模式，如ˈ green / ˈ greener , ˈ green / ˈ greenest , ˈ green / ˈ greenish , ˈ hope / ˈ hopeful , ˈ mind / ˈ mindless , ˈ slime / ˈ slimy . 当这些重音中性后缀添加到具有基本的三叉音重音模式的双音节词时，结果是倒数第二重音模式，如ˈ heavy /heavier , ˈ heaviest , ˈ heavyish , ˈ penny / ˈ penniless , ˈ pity / ˈ pitiful , ˈ summer / ˈ summery . 再一次，在派生词中可能出现音节辅音的情况下（例如bubble : [bʌbɫ̩]），添加- y时会丢失一个音节，例如 ˈ bubble / ˈ bubbly , ˈ crumble / ˈ crumbly , ˈ purple / ˈ purply , ˈ winter / ˈ wintry , ˈ wriggle / ˈ wriggly. 因此，结果是具有倒数第二重音的双音节形容词。

The native Germanic adjectival suffixes such as -er, -est, -ish, -ful, -less, -y also have no effect on stress, so when they are added to monosyllabic roots, the result is a trochaic stress pattern, as in ˈgreen/ˈgreener, ˈgreen/ˈgreenest, ˈgreen/ˈgreenish, ˈhope/ˈhopeful, ˈmind/ˈmindless, ˈslime/ˈslimy. When these stress-neutral suffixes are added to bisyllabic words with the basic trochaic stress pattern, the result is an antepenultimate stress pattern, as in ˈheavy/ˈheavier, ˈheaviest, ˈheavyish, ˈpenny/ˈpenniless, ˈpity/ˈpitiful, ˈsummer/ˈsummery. Once again, in cases where a syllabic consonant is possible in the deriving word (e.g. bubble: [bʌbɫ̩]), a syllable is lost when -y is added, as in ˈbubble/ˈbubbly, ˈcrumble/ˈcrumbly, ˈpurple/ˈpurply, ˈwinter/ˈwintry, ˈwriggle/ˈwriggly. The result is therefore a bisyllabic adjective with penultimate stress.

除了标记形容词比较级形式的 - er后缀（如greener中），还有一个重音中性的本土日耳曼语 - er后缀可以使用从动词构成名词，如ˈ advertise / ˈ advertiser , ˈ love / ˈ lover , ˈ make / ˈ maker , pre ˈ tend / pre ˈ tender , ˈ publish / ˈ publisher , ˈ sing / ˈ singer。从这些例子中我们可以看出，这个-er后缀的加入对重音没有影响。

In addition to the -er suffix which marks the comparative form of adjectives (as in greener), there is a stress-neutral native Germanic -er suffix which can be used to form nouns from verbs, as in ˈadvertise/ˈadvertiser, ˈlove/ˈlover, ˈmake/ˈmaker, preˈtend/preˈtender, ˈpublish/ˈpublisher, ˈsing/ˈsinger. We can see from these examples that the addition of this -er suffix has no effect on stress.

其他用于构成名词的本土日耳曼语后缀是 - ess、 - hood、 - ism、 - ist、 - ness和 - ship，如ˈ priest / ˈ priestess , ˈ shepherd / ˈ shepherdess , ˈ child / ˈ childhood , ˈ adult / ˈ adulthood ^, ˈ Marx / ˈ Marxism , ˈ Thatcher / ˈ Thatcherism , ˈ Marx / ˈ Marxist , ˈ Union / ˈ Unionist, ˈ capital / ˈ capitalist , ˈ kind / ˈ kindness , ˈ gentle / ˈ gentleness , ˈ friend / ˈ friendship , ˈ prentice / a ˈ prerenticeship . 再一次，添加后缀不会改变重音。

Other native Germanic suffixes, used to form nouns, are -ess, -hood, -ism, -ist, -ness, and -ship, as in ˈpriest/ˈpriestess, ˈshepherd/ˈshepherdess, ˈchild/ˈchildhood, ˈadult/ˈadulthood,4 ˈMarx/ˈMarxism, ˈThatcher/ˈThatcherism, ˈMarx/ˈMarxist, ˈUnion/ˈUnionist, ˈcapital/ˈcapitalist, ˈkind/ˈkindness, ˈgentle/ˈgentleness, ˈfriend/ˈfriendship, aˈpprentice/aˈpprenticeship. Once again, the addition of the suffix does not shift the stress.

有些拉丁语后缀无法转移重音。其中包括形成形容词的双音节后缀-able，发音为 [əbɫ̩]，如de ˈ batable 、de ˈ pendable 、 ˈ doable 、per ˈ suadable 、 ˈ sellable。我们可以看到，当-able被添加到语素中时de ˈ bate , de ˈ pend , ˈ do , per ˈ suade , ˈ sell压力不会转移。在这些情况下，结果是一个具有倒数第二个主要重音的形容词。如果将 -able添加到最后一个音节上没有主重音的单词，例如ˈ argue、 ˈ manage、 ˈ market或ˈ perish，结果是主重音放在倒数第二个音节之前的单词：可以说，“易于管理”，“适销对路”，“易腐烂。

There are Latinate suffixes which fail to shift stress. Among these are the adjective-forming bisyllabic suffix -able, pronounced [əbɫ̩], as in deˈbatable, deˈpendable, ˈdoable, perˈsuadable, ˈsellable. We can see that, when -able is added to the morphemes deˈbate, deˈpend, ˈdo, perˈsuade, ˈsell the stress does not shift. The result, in these cases, is an adjective with antepenultimate primary stress. If -able is added to a word which does not have primary stress on the final syllable, such as ˈargue, ˈmanage, ˈmarket or ˈperish the result is a word with primary stress placed prior to the ante-penultimate syllable: ˈarguably, ˈmanageable, ˈmarketable, ˈperishable.

8.5.2 重音转移后缀

8.5.2 Stress-Shifting Suffixes

在重音转移派生后缀中，我们可以区分主要重音落在的那些和那些在后缀所附的基本形式内转移重音的词尾。让我们从主要重音落在的后缀开始。

Among the stress-shifting derivational suffixes, we may distinguish between those on which the primary stress falls, and those which shift the stress within the base form to which the suffix is attached. Let us begin with suffixes on which the primary stress falls.

后缀-ee、-eer和-ese都带主重音，如em ploy /ˌ emplo ˈyee mountain /ˌ mountai neer和com puter / com ˌ pute ˈrese). 请注意，在包含其中一个后缀的多音节词中，我们不能在派生词的开头留下一系列非重读音节：必须出现次要重读。这就是 8.3 中介绍的节奏原理。推导原则和避免压力冲突原则也可能在这里发挥作用。以computer/computerese对为例：-ese后缀采用主重音，但节奏原则表明我们不能在单词开头留下一系列非重音音节的结果单词。派生原则说要将次要重音放在派生词中具有主要重音的音节上，在这种情况下，音节是动词计算的最后一个音节. 因此，压力模式compute ˌ rese 。这里没有违反重音冲突避免原则，因为次要和主要重音不落在相邻的音节上。然而，回想一下，在日语这样的词中，派生原则和重音冲突避免原则之间存在冲突。一旦我们将主要重音放在-ese后缀上，节奏原则就会坚持使用次要重音。派生原则说这个次要重音应该落在派生词Ja pan中包含主要重音的音节上. 但如果我们将次要应力放在那里，这将违反应力冲突避免原则。正如我们所看到的，当这两个原则之间存在冲突时，是避免压力冲突原则占主导地位。因此重音模式ˌJapaˈnese 。同样的情况出现在像 ˌ emplo ˈ yee这样的词上，其中次要重音没有落在主重音落在词em ˈ ploy上的地方。

The suffixes -ee, -eer and -ese all take primary stress, as in emˈploy/ˌemploˈyee ˈmountain/ˌmountaiˈneer and comˈputer/comˌputeˈrese). Notice that, in polysyllabic words containing one of these suffixes, we cannot leave the derived word with a sequence of unstressed syllables at the beginning: a secondary stress must occur. This is the Rhythmic Principle introduced in 8.3. The Derivational Principle and the Stress Clash Avoidance Principle may also play a role here. Take the pair computer/computerese: the -ese suffix takes primary stress, but the Rhythmic Principle says that we cannot leave the resulting word with a sequence of unstressed syllables at the beginning of the word. The Derivational Principle says to put that secondary stress on the syllable which had primary stress in the deriving word, in this case on the syllable which is the final syllable of the verb compute. Thus the stress pattern comˈputeˌrese. There is no violation here of the Stress Clash Avoidance Principle, since the secondary and primary stresses do not fall on adjacent syllables. Recall, however, that in a word such as Japanese, there is a clash between the Derivational Principle and the Stress Clash Avoidance Principle. Once we have placed primary stress on the -ese suffix, the Rhythmic Principle insists on a secondary stress. The Derivational Principle says that this secondary stress should fall on the syllable containing the primary stress in the deriving word Jaˈpan. But if we were to place the secondary stress there, this would violate the Stress Clash Avoidance Principle. As we have seen, when there is a conflict between those two principles, it is the Stress Clash Avoidance Principle which predominates. Thus the stress pattern ˌJapaˈnese. The same situation arises for words such as ˌemploˈyee in which the secondary stress does not fall where the primary stress falls on the word emˈploy.

我们已经论证过，许多从法语中借来的以-ette结尾的拼写单词，例如etiquette和gazette ，在形态上并不真正复杂：很难说它们包含词素etiq-和gaz-。然而，我们允许有一些词形明显复杂的-ette词，例如kitchenette一词。我们这样做的理由是这样一个词中显然有一个语素厨房，而小厨房确实是一个小厨房。鉴于我们已经允许这样做，我们可以说，至少在某些方面，有一个-ette带有主重音的后缀，因此与-ee、-eer和-ese 平行。我们看到，在-ette有明确含义的地方，它可以表示“小”（如kitchenette和sermonette）或“女性”（如ladette一词）。有些-ette词中是否有适当的-ette后缀并不完全清楚。虽然说 usherette 是女性引座员似乎是合理的，但并不清楚 maisonette 是一个小房子（法语中“房子”的意思），尽管它肯定是一个小房子。

We have argued that many words borrowed from French which have an -ette ending in the spelling, such as etiquette and gazette, are not really morphologically complex: it is hard to argue that they contain the morphemes etiq- and gaz-. However, we allowed that there are some -ette words which are clearly morphologically complex, as in the word kitchenette. Our grounds for doing so were that there is clearly a morpheme kitchen in a word like this, and a kitchenette is indeed a small kitchen. Given that we have allowed this, we can say that, at least in some words, there is an -ette suffix which takes primary stress, and is thus parallel to -ee, -eer and -ese. We saw that, where -ette has a clear meaning, it can mean either ‘little’ (as in kitchenette and sermonette) or ‘female’ (as in the word ladette). There are some -ette words in which it is not entirely clear whether there is a proper -ette suffix or not. While it might seem reasonable to say that an usherette is a female usher, it is not clear that a maisonette is a small maison (the French word for ‘house’), although it’s certainly a small house.

本身不带主重音的重音转移后缀的例子是-ity和-ic，如ˈ personal /ˌ person ˈ nality , ˌ indi ˈ vidual / ˌ individu ˈ ality , ˈ atom / a ˈtomic , ˈ monarch / mo ˈ narchic . 在每种情况下，主要重音都落在重音转移后缀之前的音节上。

Examples of stress-shifting suffixes which do not themselves take primary stress are -ity and -ic, as in ˈpersonal/ˌpersoˈnality, ˌindiˈvidual/ˌindividuˈality, ˈatom/aˈtomic, ˈmonarch/moˈnarchic. In each case, the primary stress falls on the syllable immediately preceding the stress-shifting suffix.

转移重音的后缀的其他例子是-ous（如ad ˈ vantage /ˌ advan ˈ tageous ）和-ious （如ˈ injure / in ˈ jurious ）。同样，主要重音落在重音转移后缀之前的音节上。

Further examples of suffixes which shift stress are -ous (as in adˈvantage/ˌadvanˈtageous) and -ious (as in ˈinjure/inˈjurious). Again, the primary stress falls on the syllable immediately preceding the stress-shifting suffix.

您可能已经注意到，当单词中的重音因添加重音转移后缀而发生变化时，这可能会改变受影响音节中元音的发音。因此，在personal中，最后一个音节，在-al后缀中，有一个 schwa 元音 ([ə])，而在personality中，倒数第二个音节，同样在-al后缀中，有一个 [æ] 元音，因为那个音节承担主要人格压力。添加后缀时也可能发生基本形式的辅音变化。例如，当-ity被添加到形容词opaque时，生成的形式为opacity在基数的末尾有一个 [s]，而不是 [k]。⁵此类更改不仅限于重音转移后缀。例如，当后缀-y添加到president时，生成的形式presidency在基数的末尾有一个 [s]，而不是 [t]。显然，就重音模式和辅音和元音实现而言，基本形式和附加形式之间的变化越大，基本形式和附加形式之间的实际相关性就越不明显：而之间的关系，比方说，粗体和粗体（没有重音、元音或辅音变化）是透明的，介于不透明和不透明之间远不是这样。正是由于这个原因，许多人更喜欢添加-ness类型的词缀（它往往是英语所属的日耳曼语系的原生词缀，不像-ity类型，它们的根源是拉丁语言）基本形式。

You may already have noted that, when the stress in a word shifts as a result of the addition of a stress-shifting suffix, this can have the effect of changing the pronunciation of the vowel in the affected syllable. Thus, in personal, the final syllable, in the -al suffix, has a schwa vowel ([ə]) while, in personality, the antepenultimate syllable, again in the -al suffix, has an [æ] vowel, since that syllable bears primary stress in personality. Consonantal changes in the base form may also occur when suffixes are added. For instance, when -ity is added to the adjective opaque, the resulting form opacity has an [s], rather than a [k], at the end of the base.5 Such changes are not limited to stress-shifting suffixes. For instance, when the suffix -y is added to president, the resulting form, presidency, has an [s], rather than a [t], at the end of the base. Clearly, the more variation there is, in terms of both stress pattern and consonant and vowel realization, between base form and affixed form, the less evident it will be that the base and affixed forms are actually related: while the relationship between, say, bold and boldness (with no stress, vowel or consonant changes) is transparent, that between, say, opaque and opacity is much less so. It is for this reason that many people prefer to add affixes of the -ness sort (which tend to be native to the Germanic family of languages to which English belongs, unlike the -ity sort, which have their roots in the Latinate languages) to base forms.

英语中非重读音节的一个显着特性是它们通常包含 schwa 元音，转录为 [ə]，与大多数其他元音相比，它在感知上“不太明显”。photograph和photography等相关词的一个共同特征是，当重音由于添加-y后缀而从第一个音节转移到第二个音节时，该音节中的元音从“完整”元音变为成为一个“简化”的施瓦人。我们将在下一章更详细地研究这种现象。

A striking property of unstressed syllables in English is that they often contain the schwa vowel, transcribed as [ə], which is ‘less distinct’ perceptually than most other vowels. It is a common feature of related words such as photograph and photography that, when the stress shifts from the first to the second syllable by virtue of the addition of the -y suffix, the vowel in that syllable changes from being a ‘full’ vowel to being a ‘reduced’ schwa. We will look at this kind of phenomenon in a little more detail in the following chapter.

8.5.3 单词重音模式和前缀

8.5.3 Word Stress Patterns and Prefixes

现在让我们转向前缀的重读。我们认为大多数可分离的单音节前缀都带有次要重音。所谓“可分离”，我们的意思是，如果删除前缀，我们将留下一个现有的英语单词，如动词 ˌ re -ˈ allocate , ˌ re -ˈ fabricate , ˌ re -ˈ run , ˌ re ˈ skill , ˌreˈ喷。其他单音节前缀包括：

Let us now turn to the stressing of prefixes. We will take the view that most separable monosyllabic prefixes bear secondary stress. By ‘separable’, we mean that, if the prefix is removed, we are left with an existing English word, as in the verbs ˌre-ˈallocate, ˌre-ˈfabricate, ˌre-ˈrun, ˌreˈskill, ˌreˈspray. Other monosyllabic prefixes include:

令人惊讶的是，其中一些可以用作独立词，例如“我今晚要和我的前任共进晚餐”和“你是赞成还是反对？”

It is striking that some of these can be used as independent words, as in ‘I’m having dinner with my ex tonight’ and ‘Are you with the pros or the antis?’

双音节前缀可以形成一个长音脚，因此，根据节奏原则，前缀的倒数第二个音节将承受次要重音，如ˌantiaˌbortion，ˌantibacˌterial ，ˌantiˌcatholic，ˌantiˈ choice , ˌ anticli ˈ mactic , ˌ anti - in ˈ flammatory等。虽然这些都有次要重音，但在某些情况下，双音节前缀有主要重音，例如anti-hero和antimatter. 将后面的这些情况视为复合词（由两个或更多词组成的词）也许是明智的，我们很快就会谈到。同样，虽然mega-可以有次要重音，如 ˌ mega ˈ lithic ，但在某些情况下它有主要重音，如ˈ megabyte ˈ megadeath和ˈ megaphone 。这些也可以用作独立的词，如“反派出动了”或“那部电影绝对是大片！” 其他双音节前缀是extra-，如extra-marital 、 extra-terrestrial和super-，如super-abundant 、 super-human。

Bisyllabic prefixes can form a trochaic foot, and so, in accordance with the Rhythmic Principle, will have the penultimate syllable of the prefix bearing secondary stress, as in ˌantiaˌbortion, ˌantibacˌterial, ˌantiˌcatholic, ˌantiˈchoice, ˌanticliˈmactic, ˌanti-inˈflammatory, etc. While these all have secondary stress, there are some cases where there is primary stress on bisyllabic prefixes, as in anti-hero and antimatter. It is perhaps wise to consider these latter cases as compounds (words made from two or more words), to which we turn shortly. Equally, while mega- can have secondary stress, as in ˌmegaˈlithic, there are clear cases where it has primary stress, as in ˈmegabyte ˈmegadeath and ˈmegaphone. These too may be used as independent words, as in ‘The antis are out in force’ or ‘That film was absolutely mega!’ Other bisyllabic prefixes are extra-, as in extra-marital, extra-terrestrial, and super-, as in super-abundant, super-human.

诸如刚刚列出的那些前缀都起源于拉丁语：它们出现在直接从拉丁语中借用的单词中，或者从法语中借用的单词中出现，而法语在历史上是从拉丁语中派生出来的。正如我们所见，有很多双音节名词-动词对，其中动词重音在最后一个音节上，而名词重音在拉丁语前缀上。示例是ˌdis ˈcharge （动词）vs ˈdischarge（名词），ˌex ˈport （动词）vs ˈexport（名词），ˌre ˈsearch （动词）vs ˈresearch（名词）。目前尚不清楚这些词在当代英语中在形态上仍然很复杂：例如，不能将ex - of export和 arrive at a verb port分开。但这些词肯定有一些元素，从词源上讲，这些元素是前缀。这种成对的动词在最后一个音节上重读，名词在词源前缀上重读的概括有一些例外：一些成对，例如debate、rebuke和supply，具有动词模式，而其他的，例如batter，发票和序言, 有名词模式。然而，不同的动词与名词重音模式在当代英语中似乎仍然有效。这可以从新词中看出。以ˈ vite中的动词为例：它符合模式。名词邀请已经存在了一段时间，但创造了一个更非正式的术语：一个ˈ invite。有趣的是，当新名词源自现有动词时，重音会发生变化以符合名词模式，这表明说话者可以使用名词-动词重音模式。

Prefixes such as those just listed are of Latinate origin: they occur in words which have been borrowed directly from Latin, or from words borrowed from French, which is derived historically from Latin. As we have seen, there are a good many bisyllabic noun–verb pairs in which the verb is stressed on the final syllable, whereas the noun is stressed on the Latinate prefix. Examples are ˌdisˈcharge (verb) vs ˈdischarge (noun), ˌexˈport (verb) vs ˈexport (noun), ˌreˈsearch (verb) vs ˈresearch (noun). It is not clear that such words are still morphologically complex in contemporary English: for example, one cannot separate the ex- of export and arrive at a verb port. But these words certainly have elements which are, etymologically, prefixes. There are exceptions to the generalization that the verbs in such pairs are stressed on the final syllable, and the nouns on the etymological prefix: some pairs, such as debate, rebuke and supply, have the verbal pattern, while others, such as combat, invoice and preface, have the noun pattern. However, the differential verb-vs-noun stress pattern still seems to be productive in contemporary English. This can be seen from neologisms. Take the verb inˈvite: it conforms to the pattern. The noun invitation has existed for some time, but a more informal term has been coined: an ˈinvite. Interestingly, when the new noun was derived from the existing verb, the stress shifted to conform with the noun pattern, suggesting that speakers have access to the noun-vs-verb stress patterns.

8.6 复合词

8.6 Compound Words

复合词，简而言之，就是可以被分析为由两个（或更多）词组成的词，而不是包含词基和词缀的词。例如，虽然mole-hill是一个复合词，但boldness不是（形式-ness是一个后缀，而不是一个词）。我们将在这里重点介绍两部分化合物。英语中的复合重音规则说，对于由两部分组成的化合物：在两个元素中，第一个是最突出的。因此，双词复合词与双词短语具有相反的模式，例如名词短语black bird（黑色的鸟）、形容词短语very tall、动词短语如kissed Mary、副词短语如很慢，介词短语如变成London。这些都体现了英语短语重音规则（我们将在第 9 章中返回）：在所有这些中，第二个元素是最突出的。具有规则复合应力模式的化合物的例子，第一个元素最突出，有：原子弹，背景，黑鸟，停车场，教室，复出，开瓶器，暗室，蜻蜓，文件柜，花坛，花盆、文法学校、握手、高中、化妆、地名、社交生活、性生活、暴风雪、汽船、教科书、啄木鸟。

Compound words are, put simply, words which can be analysed as consisting of two (or more) words, rather than as containing a base and an affix. For instance, while mole-hill is a compound, boldness is not (the form -ness is a suffix, not a word). We will focus on two-part compounds here. The Compound Stress Rule in English says, of two-part compounds: of the two elements, the first is the most prominent. Two-word compounds thus have the opposite pattern to two-word phrases, such as the noun phrase black bird (a bird which is black), the adjective phrase very tall, verb phrases such as kissed Mary, adverb phrases such as very slowly and prepositional phrases such as into London. These all exhibit the English Phrasal Stress Rule (which we will return to in chapter 9): in all of these, it is the second element which is most prominent. Examples of compounds which have the regular compound stress pattern, with the first element the most prominent, are: atom-bomb, backdrop, blackbird, car-park, classroom, comeback, corkscrew, darkroom, dragonfly, filing cabinet, flower-bed, flowerpot, grammar school, handshake, high-school, make-up, place-name, social life, sex life, snowstorm, steamboat, textbook, woodpecker.

我们如何判断给定的双词序列是复合词还是短语？当两个部分写成一个词（例如flowerpot）或连字符（例如atom-bomb）时，很容易看出这是在处理一个化合物。但是如果这两个部分是作为单独的词来写的（例如grammar school），那就不太容易了。（有复合词的书写方式也有所不同：例如，人们可能会发现书面形式textbook、text-book 和 text book。）在许多情况下，复合词具有与短语不同的语义（含义）。以短语black bird、dark room和green house为例。将它们的含义与化合物blackbird、darkroom和greenhouse进行比较. 虽然所有（雄性）黑鸟都是黑鸟，但并非所有黑鸟（短语）都是黑鸟（复合）：乌鸦、寒鸦和鸬鹚是黑鸟（短语），但它们不是黑鸟（复合）。虽然所有的暗房都是通常黑暗的房间，但并非所有的暗房都是暗房（冲洗照片的地方）：如果我关上百叶窗并关掉书房的灯，它就会变成暗房（短语），但不是暗房（化合物）。A green house (phrase) 是涂成绿色的房子，但 greenhouse (compound) 不是房子，可以涂成白色。在英语的历史上，复合词似乎开始于短语：一只（雄性）黑鸟确实是黑色的，一间暗室通常确实是黑暗的，而温室是一种房屋形状的结构，人们可以在其中种植绿色植物。

How can we tell whether a given two-word sequence is a compound or a phrase? When the two parts are written as one word (e.g. flowerpot) or with a hyphen (e.g. atom-bomb), it is easy to see that one is dealing with a compound. But if the two parts are written as separate words (e.g. grammar school), it is less easy. (There is also variation in how compounds are written: one may find, for example, the written forms textbook, text-book and text book.) In many cases, compounds have a different kind of semantics (meaning) from phrases. Take the phrases black bird, dark room and green house. Compare their meanings with the compounds blackbird, darkroom and greenhouse. While all (male) blackbirds are black birds, not all black birds (phrase) are blackbirds (compound): ravens, jackdaws and cormorants are black birds (phrase), but they are not blackbirds (compound). While all darkrooms are rooms which are normally dark, not all dark rooms are darkrooms (places for developing photographs): if I close the shutters and switch off the lights in my study, it becomes a dark room (phrase), but not a darkroom (compound). A green house (phrase) is a house which is painted green, but a greenhouse (compound) is not a house, and may be painted white. It seems likely that, in the history of English, compounds started off as phrases: a (male) blackbird is indeed black, a darkroom is indeed normally dark, and a greenhouse is a house-shaped structure where one grows green things. But such phrases have made the transition to becoming single words.

8.6.1 复合应力规则的例外

8.6.1 Exceptions to the Compound Stress Rule

(a) 由两部分组成的地名，例如Botany Bay、Buckingham Palace、East Anglia、Los Angeles、Mount Everest、New York、Niagara Falls、San Francisco。这些包括街道名称，例如黑莓路、第五大道、伦敦路、莫宁顿新月、穆赫兰道、便士巷、佩顿广场、特拉法加广场。唯一的例外是大量街道名称以单词street结尾：在我的家乡爱丁堡，London Road（Road最显眼的地方）离 London Street（London最显眼的地方）不远。

有些化合物系统地违反了复合应力规则。我们现在列出这些。

(b) 具有分词第二元素的化合物，以-ed、-en或-ing结尾。-ed化合物比后者更常见。其中许多是基于身体的各个部分，其中一些比其他的更具隐喻性。例子是：ˌ光秃秃的-面对，ˌ大-有耳朵，ˌ大-有头，ˌ破碎-有心，ˌ cack -有手，ˌ昏暗-机智，ˌ双-有关节，ˌ空-有头, ˌ even -ˈ handed ˌ even -ˈ脾气暴躁ˌ weak -ˈ hearted ˌ fair -ˈ haired ˌ far -ˈ sighted ˌ flat -ˈ chested ˌ flat -ˈ footed ˌ fleet -ˈ footed , ˌ freet -ˈ footed , ˌ foul -ˈ mouthed ˌ good -ˈ natured ˌ hard -ˈ nosed ˌ high -ˈ pitched ˌ hot -ˈheaded ˌ ill - di sposed ˌ ill -脾气暴躁ˌ kind - hearted ˌ left - handed ˌ level - headed ˌ lily - livered ˌ limp -手腕ˌ long - haired , ˌ long -ˈ legged ˌ long -ˈ winded , ˌ narrow -ˈ shouldered , ˌ old -ˌ时尚，ˌ一个-眼睛， ˌ一个-有腿，ˌ猪-有头，红-手，红-头，吱吱作响-声音，强壮-头脑，强壮-意志， ˌ thin -ˈ lipped ˌ weak -ˈ kneed ˌ weak -ˈ willed ˌ wrong -ˈ footed , ˌ wrong-有头脑，ˌ干净-刮胡子， ˌ长-成熟， ˌ柔和-口语， ˌ清晰-思维， ˌ深远-深远， ˌ漂亮-长相。

(c) 复合词的第一部分表示物体是由什么制成的。例如：苹果派、砖墙、棉袜、火腿三明治、铁屑、餐巾纸、猪肉派、橄榄油。请注意，这些与类似的化合物不同，在这些化合物中，第一部分不表示物体是由什么制成的，例如回形针不是纸制成的，棉卷不是棉花制成的，橄榄树不是用棉花制成的。不是用橄榄制成的。请注意，在美式英语中，第一个元素可能是此类复合词中最突出的。

(d) 涉及某种具体或抽象定位的化合物。例如：四月的阵雨、圣诞假期、晚餐、中产阶级、二流、寒假。

(e) 由两部分组成的颜色词的复合词。例如：深绿色、深黄色、浅绿色、淡蓝色。

(f) 源自短语动词的复合词。示例是从短语动词派生的复合名词：chucker-out（来自chuck out）、hanger-on（来自hang on）、passer-by（来自pass by）、washing-up（来自wash up）；和来自短语动词的复合形容词：finished-off (from finish-off ) , knocked- out , pared -down (from pare down ), rolled-up (from roll up ), tired-out, wiped-out , wrapped-向上（从总结）。

8.7 总结

8.7 Summing Up

我们在本章中介绍了相当多的细节。此处总结要点可能对读者有帮助，忽略许多细节和例外情况：如果读者能够掌握本章涵盖的每个要点的关键信息，则可以通过查阅每个要点来掌握详细信息本章的一部分。

We have covered a fair amount of detail in this chapter. It may prove helpful to the reader to have the main points summarized here, ignoring many details and exceptions: if the reader can grasp the take-home message for each of the points covered in this chapter, the details can then be mastered by consulting each section of the chapter.

• 英文单词stress 不是随机的。

• 英语节奏是有节奏的，如woman和battery。

• 主重音是从单词的结尾而不是开头计算的。

• 英语单词不能以一个以上的非重读音节开头。

• 当一个英语单词派生自另一个单词，并且主重音因派生而发生变化时，会倾向于将次重音放在派生词中具有主重音的音节上，如 ˌ characteri ˈ zation 。

• 人们倾向于避免将主要重音和次要重音放在一起，就像在ˌ Japa ˈ nese中一样。

• 虽然英语名词、形容词和副词主要遵循基本的三义词模式，但也有许多动词不遵循。

• 英语后缀可分为影响主重音的后缀（例如-ese和-ity）和不影响主重音的后缀（例如-ness）。

• 在影响首重音的后缀中，有的带首重音（如-ese），有的则不带首重音（如-ity）。

• 可分离前缀通常带有次要重音，例如ˌpre -ˈpay。

• 由两部分组成的化合物的基本模式是：第一个元素是最突出的，就像在暗室中一样。

笔记

Notes

1如果任何这样的结尾可能是富有成效的，并因此可能导致新词，那么由此产生的词确实在形态上是复杂的。例如，最近创造的词ladette（在英国，一个女小伙子：一个行为吵闹、满嘴脏话、酗酒、性滥交的年轻女性）在形态上很复杂（这里的-ette意思是“女性”，而不是“小”）。

1 To the extent that any such endings may be productive, and thus may result in neologisms, then the resulting words will indeed be morphologically complex. For instance, the fairly recently coined word ladette (in Britain, a female lad: a young woman who behaves in a loud, foul-mouthed, heavy-drinking, sexually promiscuous manner) is morphologically complex (here -ette means ‘female’, rather than ‘little’).

2考虑到cosmos这个词的存在，有人可能会争辩说cosmic这样的词在形态上是复杂的。我们认为cosm - 不构成现代英语中的语素，因此cosmos和cosmic等词在形态上是简单的。

2 It might be argued that words such as cosmic are morphologically complex, given the existence of the word cosmos. We take the view that cosm- does not constitute a morpheme in contemporary English, and that words such as cosmos and cosmic are therefore morphologically simple.

3然而，有理由说kitchenette这样的词在形态上是复杂的，因为kitchen显然是一个语素，而 kitchenette 确实是一个小厨房。

3 There is, however, a case for saying that words such as kitchenette are morphologically complex, since kitchen is clearly a morpheme, and a kitchenette is indeed a small kitchen.

4请注意，adult通常在美式英语的最后一个音节上重读。

4 Note that adult is typically stressed on the final syllable in American English.

5 opacity这个词在 [eɪ] 和 [æ] 之间的重读元音中也表现出交替。这是我们在这里不检查的一组元音交替中的一个。

5 The word opacity also exhibits an alternation, between [eɪ] and [æ], in the stressed vowel of the base. This is one of a set of vowel alternations which we do not examine here.

6表达bouquet garni（烹饪中使用的一束或一袋药草）的主重音在bouquet的倒数第二个音节上，而不是最后一个音节上。这是抑扬格反转过程的结果，稍后将在我们对韵律结构的讨论中进行描述。

6 The expression bouquet garni (a bunch or sachet of herbs used in cooking) is pronounced with primary stress on the penultimate, rather than the final, syllable of bouquet. This is a result of the process of iambic reversal, described later in our discussion of metrical structure.

练习

Exercises

gueqrecard_001

1在www.wiley.com/go/carrphonetics收听曲目 8.1。对于录音中的每个双音节词，说出哪些具有默认的主重音的次级重音模式。对于那些偏离该模式的人，请解释原因。话是：

（饥荒

(b) 马耳他人

(d) 实习生

(e) 冬天

(f) 解释

(七) 傻

(h) 契约（名词）

(i) 紧凑（动词/形容词）

(j) 出口（名词）

(k) 出口（动词）

(l) 绊倒

(m) 花式

(n) 不同

(o) 禁忌

(p) 公报

(q) 街机

(r) 滑稽表演

2 听曲目 8.2。对于录音中的每组多音节词，说出它们具有默认的日耳曼语托查语模式。对于那些偏离默认模式的，解释原因。套数如下：

（一）工厂

美国

家庭

学院

刺激计划

(b) 发展

继承

同谋

显式

继承

(d) 讯问

调查

容纳

证明

坐标

(e) 数学

物理

周期性

语言学

酗酒

(f) 秋日

感伤

管弦

横式

通用

(g) 敌意

紧缩

现代性

谦逊

模棱两可

(h) 香蕉

比基尼

空手道

马提尼

钢琴

(i) 冒险

混合

共识

11 月

优势

(j) 临时

秘书

文学

实验室

军事

gueqrecard_002

3 听曲目 8.3。录音中的每个名词（都是英语中的外来词）都有一种不同于默认日耳曼语模式的重音模式。在每种情况下，说出在什么意义上重音模式相对于英语单词重音分配规则而言是例外的。主要和次要重音标记如下：[ˌhəʊˈtεɫ]（hotel，主要重音在最后一个音节上，次要重音在倒数第二个音节上）。在 GA 和 RP 发音不同的地方，会指出这一点，如 [ˌhəʊˈtεɫ]/[ˌhoʊˈtεɫ]，它给出 RP 发音后跟 GA 发音。

（一种）	酒店	（[ˌhəʊˈtεɫ]/[ˌhoʊˈtεɫ]，而不是 [ˌhəʊtεɫ]/[ˌhoʊtεɫ]）
（乙）	花束	([ˌbʊˈkʰeɪ], 不是 [ˈbu:keɪ]) ⁶
（C）	竹子	（[ˌbæmˈbu:]，而不是 [ˌbæmbu:]）
(四)	香槟酒	（[ˌʃæmˈpʰeɪn]，而不是 [ˈʃæmpeɪn]）
(五)	比基尼	（[bɪkʰi:ni]，而不是 [bɪkʰini]）
（F）	马提尼酒	（[mɒ:ˈtʰi:ni]/[mɒɹˈtʰi:ni]，而不是 [ˈmɒ:tini]/[ˈmɒɹtini]）
（G）	香肠	（[tʃəˈɹi:zoʊ]，而不是 [ˈtʃɑɹɪzoʊ]）

gueqrecard_002

4 对于非母语人士：听曲目 8.4并重复每个话语。对于以母语为母语的人和非母语人士：在下面给出的这些话语的转录中指出主要压力和次要压力。例如：

玛丽发现比尔的书无法解释。ˈmeəɹi ˈfaɪndz ˈbɪɪz ˈbʊk ˌʌnɪnˈtɜ:pɹətəbɪ̩

（一种）	数学是非常困难的。mæθəmætɪksɪzɪŋkɹεdɪbli dɪfɪkəɫt
（乙）	我的车是美国制造的。maɪ kɒ: wəs meɪd ɪn əmεɹɪkə
（C）	他的电脑是日本的。hɪz kəmpju:təɹɪz dʒæpəni:z
(四)	学术对话很乏味。ækədemɪk kɒnvəseɪʃn̩ɪz dʌɫ
(五)	警方将审问被拘留者。ðə pəli:s wɪɪ ɪntεɹəgeɪt ðə diteɪni:z
（F）	他们不生产很多出口产品。ðeɪ dəʊnt pɹədju:s mεni εkspɔ:ts
（G）	他们出口的产品不多。ðeɪ dəʊnt ɪkspɔ:t mʌtʃ pɹɒdju:s
（H）	我觉得这部电影很伤感。aɪ faʊnd ðə fɪɪm ɹɒ:ðə sεntɪmεntɫ̩

gueqrecard_002

5 对于非母语人士：听曲目 8.5并复述你听到的单词。对于以英语为母语的人和非母语人士一样：解释录音中听到的以下英语双音节词组的单词重音模式。

(a) 碰巧

女人

喜欢

回声

父亲

(b) deny

inspect

compliance

展开

变形

（三）实习生

竹

芭莎

禁忌

洗发水

(d) 创建

迁移

定位

挫败

叙述

(e) 生产（动词）

出口（动词）

排放（动词）

对象（名词）

合同（名词）

(f) produce（名词）

export（名词）

discharge（动词）

object（名词）

contract（名词）

gueqrecard_002

6 听曲目 8.6。如录音中所听到的，解释以下句子中的格律英尺边界在哪里。（在脚边界落下的地方画垂直线。）

（一）放开我！

(b) 给我一片！

(d) 她住在美国。

(e) 放入冰箱！

(f) 约翰是现代型男。

(g) 克林顿反对军国主义解决方案。

(h) Aude 是一个弹性素食主义者。

9 节奏、反转和还原

Rhythm, Reversal and Reduction

9.1 关于 Trochaic 公尺的更多信息

9.1 More on the Trochaic Metrical Foot

我们在第 8 章中说过，英语的节奏是trochaic：基本的节奏模式由一个重读音节后面跟着零个或多个非重读音节组成。例如，在made in a factory短语中，韵律结构是 [ˈmeɪdɪnəˈfæktəɹi]。这里的两个长音脚是 [ˈmeɪdɪnə] 和 [ˈfæktəɹi]。我们还假设具有次要重音的音节也形成了 trochaic 格律脚，如学术一词：[ˌækədεmɪk]。这里的两个韵律脚是 [ˌækə] 和 [ˈdεmɪk]：[ˌækə] 中的次重音与后面的非重读音节形成一个韵律韵脚，[dεmɪk] 中的主重音与下面的非重读音节构成一个韵律韵脚音节。

We said, in chapter 8, that the rhythm of English is trochaic: the basic rhythmic pattern consists of a stressed syllable followed by zero or more unstressed syllables. For instance, in the phrase made in a factory, the metrical structure is [ˈmeɪdɪnəˈfæktəɹi]. The two trochaic feet here are [ˈmeɪdɪnə] and [ˈfæktəɹi]. We assumed too that syllables with secondary stress also form trochaic metrical feet, as in the word academic: [ˌækəˈdεmɪk]. The two trochaic metrical feet here are [ˌækə] and [ˈdεmɪk]: the secondary stress in [ˌækə] forms a trochaic metrical foot with the following unstressed syllable, and the primary stress in [ˈdεmɪk] forms a trochaic metrical foot with the following unstressed syllable.

但是，米尺的证据是什么？有什么证据可以证明我们声称英语中的所有脚都是 trochaic？我们现在将解决这些问题。

But what is the evidence for the metrical foot? And what evidence is there for our claim that all feet in English are trochaic? We will now address these questions.

9.1.1 Trochaic 韵脚的证据 (a)：押韵

9.1.1 Evidence for the Trochaic Metrical Foot (a): Rhyming

虽然我们已经确定了音节中广为人知的押韵成分，但该术语用词不当：该成分不是英语押韵所基于的单位。虽然bad确实与mad押韵，并且两者都包含押韵 [æd]，但我们绝不能误以为两个词只有在音节中具有相同的押韵成分时才押韵，在本例中为 [æd]。想一想witty和city这两个词：它们押韵是因为它们都有韵律。同类脚：[ˈwɪti] 和 [ˈsɪti]。显然，起始辅音在押韵中不起作用，但韵律结构起作用，而押韵成分则不起作用。entity之所以不与witty或city押韵，是因为entity的韵脚结构是[ˈεntɪti]：entity一词不含[ˈɪti] 形的韵脚。英语中的押韵是基于两个或更多的韵律韵脚中重读元音的同一性，以及所有后续语音片段的同一性。¹ entity这个词与witty和city两个词没有相同的重读元音。

Although we have identified a constituent within the syllable widely known as the rhyme, the term is a misnomer: this constituent is not the unit on which rhyming in English is based. While it is true that bad rhymes with mad, and that both contain the rhyme [æd], we must not be misled into thinking that two words rhyme only if they have identical rhyme constituents in the syllable, in this case [æd]. Consider the words witty and city: they rhyme because they both have a trochaic metrical foot of the same sort: [ˈwɪti] and [ˈsɪti]. Clearly, onset consonants play no role in rhyming, but metrical structure does, and the rhyme constituent does not. The reason why entity does not rhyme with either witty or city is that the metrical foot structure of entity is [ˈεntɪti]: the word entity does not contain a metrical foot of the shape [ˈɪti]. Rhyming in English is based on identity of stressed vowels in two or more trochaic metrical feet, and identity of all subsequent phonetic segments.1 The word entity does not have the same stressed vowel as the words witty and city.

同样，短语phone ya ( phone you ) 在 RP 中发音为 [ˈfəʊnje]，在 GA 中发音为 [ˈfoʊnje]，与pneumonia押韵：在 RP 中发音为 [nju:ˈməʊnje]，在 GA 中发音为 [nu:moʊnje]。在这种情况下，押韵是基于不直接映射到词边界的 trochaic 韵脚：[ˈfəʊnje] 包含两个词，而 [ˈməʊnje] 是一个词的一部分。

Similarly, the phrase phone ya (phone you), which is pronounced [ˈfəʊnje] in RP and [ˈfoʊnje] in GA, rhymes with pneumonia: [nju:ˈməʊnje] in RP and [nu:ˈmoʊnje] in GA. In this case, the rhyme is based on trochaic metrical feet which do not map directly onto word boundaries: [ˈfəʊnje] contains two words, while [ˈməʊnje] is part of a word.

9.1.2 Trochaic 度量脚的证据 (b)：咒语插入

9.1.2 Evidence for the Trochaic Metrical Foot (b): Expletive Insertion

许多人在非正式英语口语中经常使用bloody和f * cking等咒骂词。每个都包含一个标准的日耳曼语双音节韵律韵律脚：[ˈblʌdi] 和 [ˈfʌkɪŋ]（许多说英语的人也用词尾 [n] 发音，这可能是音节的：[ˈfʌkn̩]）。

Expletives such as bloody and f*cking are frequently used in informal spoken English by many speakers. Each consists of a standard Germanic bisyllabic trochaic metrical foot: [ˈblʌdi] and [ˈfʌkɪŋ] (also pronounced with a final [n] by many speakers of English, which may be syllabic: [ˈfʌkn̩]).

与普遍的看法相反，在句法结构中使用这些词的模式非常复杂。我们将忽略句法的复杂性，重点关注它们可以插入单词的内部结构这一事实，如在abso-bloody-lutely 中，bloody被插入单词absolutely中。这个词是从形容词absolute派生出来的副词，具有英语多音节词的基本倒数第二重音模式：[ˈæbsəlu:t]。然而，有一个强调的发音，带有最后的主要压力和次要压力在倒数：[ˌæbsəˈlu:t]。正是在这种强调形式中，可以插入咒语，例如：

The patterns of use of words such as these in syntactic structure is, contrary to popular belief, quite complex. We will ignore that syntactic complexity and focus here on the fact that they can be inserted into the internal structure of words, as in abso-bloody-lutely, where bloody is inserted into the word absolutely. This word is an adverb derived from the adjective absolute, which has the basic antepenulti-mate word stress pattern of polysyllabic words in English: [ˈæbsəlu:t]. There is, however, an emphatic pronunciation with final primary stress and a secondary stress on the antepenult: [ˌæbsəˈlu:t]. It is in this emphatic form that the expletive can be inserted into, as in:

A：你喜欢艾米·怀恩豪斯吗？

B: Abso-bloody-lutely！

如果您的母语是英语（有些非母语人士也会看到这一点），您会同意不可能回复ab-bloody-solutely或absolute-bloody-ly。但为什么不呢？答案在于有余弦韵脚的存在。咒骂词是双音节的长音韵律脚。可以插入它们的单词将包含 trochaic 韵律脚。在插入咒语时，必须尊重正在插入咒语的单词的三段式韵律结构：abso-bloody-lutely包含一系列三个三段式双音节韵律脚。它尊重绝对音节的双音节韵脚结构，在这两者之间插入另一个这样的脚。

If you are a native speaker of English (and some non-native speakers will see this too), you will agree that it is not possible to reply ab-bloody-solutely, or absolute-bloody-ly. But why not? The answer lies in the existence of the trochaic metrical foot. The expletives are bisyllabic trochaic metrical feet. The words into which they can be inserted will contain trochaic metrical feet. In inserting expletives, one must respect the trochaic metrical structure of the word one is inserting an expletive into: abso-bloody-lutely contains a sequence of three trochaic bisyllabic metrical feet. It respects the trochaic bisyllabic metrical foot structure of absolute, inserting another such foot in between those two.

9.1.3 Trochaic 度量脚的证据 (c)：新词

9.1.3 Evidence for the Trochaic Metrical Foot (c): Neologisms

许多新词都是基于 trochaic 韵律脚。这里有一些例子：酗酒者、工作狂、购物狂、性狂、巧克力狂。所有这些新词都是基于与一个或多个先前存在的词的类比。如果我们假设工作狂、购物狂、性狂、巧克力狂都基于与酒精这个词的类比，那么我们必须问类比的基础是什么。在这种情况下，它不可能是单词 alcohol 的形态结构，它包含词根alcohol和后缀-ic（正如我们所见，它转移了单词alcohol的重音, 发音为 [ˈæɫkəhɒɫ])。如果形态学是类比的基础，那么新词将是workic、shopic、sexic和chocic。

Many neologisms are based on the trochaic metrical foot. Here are some examples: alcoholic, workaholic, shopaholic, sexaholic, chocaholic. All of these neologisms are based on analogy with one or more previously existing words. If we assume that work-aholic, shopaholic, sexaholic, chocaholic are all based on an analogy with the word alcoholic, then we must ask what the basis of the analogy might be. In this case, it cannot be the morphological structure of the word alcoholic, which contains the root alcohol and the suffix -ic (which, as we have seen, shifts the stress from the word alcohol, pronounced [ˈæɫkəhɒɫ]). If the morphology were the basis for the analogy, then the neologisms would be workic, shopic, sexic and chocic.

这些情况下的类比是基于单词alcoholic ([ˌælkəˈhɒlɪk]) 的韵律结构，它包含两个双音节长音韵律脚：[ˌælkə] 和 [ˈhɒlɪk]。work、shop、sex和choc等形式必须用 <a> 书写的原因是，这个字母代表一个 schwa 元音 ([ə])，它出现在alcoholic的第一个三元韵律脚中，类比基于。以英语为母语的人会产生这样的新词，因为他们对母语的韵律结构有一种（也许不是完全有意识的）感觉。

The analogy in these cases is based on the metrical structure of the word alcoholic ([ˌælkəˈhɒlɪk]), which contains two bisyllabic trochaic metrical feet: [ˌælkə] and [ˈhɒlɪk]. The reason why the forms work, shop, sex and choc have to be written with an is that this letter represents a schwa vowel ([ə]) which is present in the first trochaic metrical foot of alcoholic, on which the analogy is based. Native speakers of English produce such neologisms because they have a (perhaps not entirely conscious) sense of the metrical structure of their native language.

想想当代的新词，例如flexitarian（准备灵活的素食主义者，不时吃肉）和pescitarian（不吃肉，但吃鱼或一般海鲜的人）。这些词是通过类比单词vegetarian形成的，在 RP 中发音为 [ˌvεdʒəˈteəɹiən]。这个词的两个韵律脚是 [ˌvεdʒə] 和 [ˈteəɹiən]。正是这种韵律结构驱动了类比过程，导致了弹性素食主义者和鱼素主义者：[ˌflεksɪˈteəɹiən] 和 [ˌpεksɪˈteəɹiən]。^2个

Consider contemporary neologisms such as flexitarian (a vegetarian who is prepared to be flexible, and eat meat from time to time) and pescitarian (someone who doesn’t eat meat, but who does eat fish, or seafood in general). These words have been formed by analogy with the word vegetarian, pronounced [ˌvεdʒəˈteəɹiən] in RP. The two trochaic metrical feet in this word are [ˌvεdʒə] and [ˈteəɹiən]. It is this metrical structure that drives the analogical process which results in flexitarian and pescitarian: [ˌflεksɪˈteəɹiən] and [ˌpεksɪˈteəɹiən].2

还要考虑最近出现的新词metrosexual（都市异性恋男性，过分关注自己的外貌）。在这种情况下，韵律脚扮演了一个角色：异性恋这个词发音为 [ˌhεtɹəʊˈsεkʃəə]。基于 trochaic 韵律脚的押韵过程在这里再次发挥作用：说话者知道异形词素可能是双音节词素 ([ˌhεtɹəʊ])，词素性词素([ˈsεkʃəə]) 也是如此。^3个

Consider too the recent neologism metrosexual (a metropolitan heterosexual man who is overly concerned with his physical apearance). In this case, the trochaic metrical foot plays a role: the word heterosexual is pronounced [ˌhεtɹəʊˈsεkʃəɫ]. The rhyming process, based on the trochaic metrical foot, is again at work here: speakers know that the hetero morpheme may be bisyllabic ([ˌhεtɹəʊ]), as is the morpheme sexual ([ˈsεkʃəɫ]).3

最近的品牌名称safetergent（清洁产品）基于洗涤剂一词：[dɪˈtʰɜ:dʒənt] (RP)、[dɪˈtʰɜɹdʒənt] (GA)。这个词的脚结构有一个初始的非韵律非重读音节，留下 [ˈtʰɜ:dʒənt] (RP)/ [ˈtʰɜɹdʒənt] (GA) 作为长音韵律脚。然后我们可以将语素安全([seɪf]) 添加到这个韵律脚以产生safetergent : [ˌseɪfˈtʰɜ:dʒənt] (RP)/ [ˌseɪfˈtʰɜɹdʒənt] (GA)。

The recent brand name safetergent (a cleaning product) is based on the word detergent: [dɪˈtʰɜ:dʒənt] (RP), [dɪˈtʰɜɹdʒənt] (GA). The foot structure of this word has an initial extrametrical unstressed syllable, leaving [ˈtʰɜ:dʒənt] (RP)/ [ˈtʰɜɹdʒənt] (GA) as the trochaic metrical foot. We can then add the morpheme safe ([seɪf]) to this metrical foot to produce safetergent: [ˌseɪfˈtʰɜ:dʒənt] (RP)/ [ˌseɪfˈtʰɜɹdʒənt] (GA).

同样，单词kissagram（[ˈkɪsəgɹæm]：用吻传递电报类型信息的人）和stripagram（[ˈstɹɪpəgɹæm]：传递此类消息和条带的人）是通过类比电报一词形成的，其韵律结构符合英语中多于两个音节的单词的基本三叉音结构，因为它具有倒数第二个主重音：[ˈtʰεləgɹæm]。虽然telegram的形态结构是tele + gram，但新词不是kissgram和stripgram。这样做的原因是新词是仿照电报的韵律结构。英语的 trochaic 韵律脚显然在我们在这里考虑的所有新词中都扮演着重要角色。

Similarly, the words kissagram ([ˈkɪsəgɹæm]: someone who delivers a telegram-type message with a kiss) and stripagram ([ˈstɹɪpəgɹæm]: someone who delivers such a message and strips) are formed by analogy with the word telegram, whose metrical structure conforms to the basic trochaic structure for words of more than two syllables in English in that it has antepenultimate primary stress: [ˈtʰεləgɹæm]. Although the morphological structure of telegram is tele + gram, the neologisms are not kissgram and stripgram. The reason for this is that the neologisms are modelled on the metrical structure of telegram. The English trochaic metrical foot clearly plays a role in all of the neologisms we have considered here.

9.2 表示韵律结构

9.2 Representing Metrical Structure

我们用上标变音符号表示主要重音，如 [ˈdεmɪk]，并用下标变音符号表示次要重音，如单词academic中的 [ˌækə] 。如果我们将兴趣限制在单词的层面上，这些约定就足够了。但是，如果我们希望表示当单词组合成短语时压力水平和相对感知显着性的运作方式，它们是不够的。以短语kangaroo court为例。当kangaroo出现在短语kangaroo court中时，该词的次要和主要重音会轮换。court这个词的单音节比kangaroo中的任何一个重读音节都重读，但是在这个词中袋鼠，[kʰæŋgə] 脚不如 [ɹu:] 脚突出。所以我们在这里处理三个不同层次的显着性。这并不容易仅使用我们用于单词重音的两个变音符号来表示。我们需要另一种表示方式。

We have represented primary stress with a superscript diacritic, as in [ˈdεmɪk], and secondary stress with a subscript diacritic, as in [ˌækə] in the word academic. These conventions will suffice if we confine our interest to the level of the word. But they will not suffice if we wish to represent the way levels of stress and relative perceptual salience operate when words are combined into phrases. Take the phrase kangaroo court, for instance. When kangaroo appears in the phrase kangaroo court, the secondary and primary stresses in that word switch round. The single syllable of the word court has more stress than any of the stressed syllables in kangaroo, but in the word kangaroo, the [kʰæŋgə] foot is less salient than the [ɹu:] foot. So we are dealing here with three different levels of salience. That is not easy to represent using only the two diacritics we have used for word stress. We need a further mode of representation.

我们用分支树结构表示音节结构。许多音韵学家还用分支树来表示足部结构。我们将用“S”表示具有任何重音程度的任何音节，表明它相对于我们标记为“W”的弱无重音音节而言是强音节。一个重读音节和任何与它构成一个脚的非重读音节可以表示如下：⁴

We represented syllable structure in terms of branching tree structures. Many phonologists also represent foot structure in terms of branching trees. We will represent any syllable which has any degree of stress with an ‘S’, indicating that it is strong with respect to weak unstressed syllables, which we label with a ‘W’. A stressed syllable and any unstressed syllables with which it forms a foot may then be represented as follows:4

(1)

此图中最底层的表示级别是段的级别。下一个级别是音节。在那个级别，S 标签代表强（重读）音节，W 标签代表非重读音节。重要的是要记住重音水平是相关的：重音音节不是用绝对术语来定义的，而是一个音节相对于另一个音节或多或少地重音。

The bottom-most level of representation in this diagram is the level of the segment. The next level up is the syllable. At that level, the S labels represent strong (stressed) syllables and the W labels represent unstressed syllables. It is important to bear in mind that stress levels are relational: rather than a stressed syllable being definable in absolute terms, one syllable is more or less stressed in relation to another.

从音节开始的下一层表示（图中 S 和 W 标签上方的线）是脚的表示。(1)中的每个字都由一个脚组成，第一个字由一个二叉脚组成，第二个字由一个三叉脚组成。

The next level of representation up from the syllable (the lines above the S and W labels in this diagram) is that of the foot. Each word in (1) consists of a single foot, the first word consisting of a binary-branching foot and the second word consisting of a tertiary-branching (three-way-branching) foot.

根据定义，单音节词包含一个重读音节。我们将认为它们包含一个单一的、非分支的脚。我们将因此将这样的词表示为具有单个 S 标记的音节（表示它是重读的），由非分支脚节点控制，因此：

Monosyllabic lexical words contain, by definition, a single stressed syllable. We will take it that they contain a single, non-branching foot. We will therefore represent such words as having a single S-labelled syllable (indicating that it is stressed), dominated by a non-branching foot node, thus:

(2)

该图代表两个声明。第一个是所讨论的音节是重读的（标记为 S）。第二个是，因为它是重读的，所以它构成了一个恰好没有分支结构的脚，因为它后面没有非重读音节。

This diagram represents two claims. The first is that the syllable in question is stressed (labelled S). The second is that, because it is stressed, it constitutes a foot which happens not to have a branching structure, since there are no unstressed syllables following it.

单音节非词汇功能词，例如代词（例如他、她、我、它）、介词（例如in、on、at）、冠词（a、the）和连词（例如and、but、if）通常是无重读的. 因此，我们将用 W 标记的音节来表示它们，但在该级别之上没有脚结构（因为根据定义，脚必须包含重读音节），因此：

Monosyllabic non-lexical function words, such as pronouns (e.g. he, she, me, it), prepositions (e.g. in, on, at), articles (a, the) and conjunctions (e.g. and, but, if) are typically unstressed. We will therefore represent them with a W-labelled syllable, but no foot structure above that level (since a foot by definition must contain a stressed syllable), thus:

(3)

在同时具有主重音节和次重音音节的词中，例如woodpecker ，可以清楚地看出重音水平的相关性。很明显，这个词中的倒数第二个音节比倒数第二个音节重音更多（相对于倒数第二个音节来说它更重）。同样清楚的是，最后一个音节的重音比倒数第二个音节小：它没有重音，因此很弱。因此，我们可以将单词的脚结构表示为由两只脚组成，第一只脚比第二只脚强壮，如下所示：⁵

The relational nature of stress levels can be seen clearly in words which have both a primary stressed syllable and a secondary stressed syllable, such as woodpecker. It is clear that the antepenultimate syllable in this word has more stress than the penultimate (it is strong with respect to the penult). It is equally clear that the final syllable has less stress than the penultimate: it is unstressed, and thus weak. We may therefore represent the foot structure of the word as consisting of two feet, the first of which is stronger than the second, as follows:5

(4)

请注意，S/W 符号用于表示一个脚内的音节和脚序列的相对强度：该符号表明第一个音节相对于第二个音节强，并且还表明，在更高的水平上，第一只脚比第二只脚强。

Note that the S/W notation is used to represent the relative strength both of syllables within a foot and of sequences of feet: the notation shows that the first syllable is strong with respect to the second, and also shows, at a higher level, that the first foot is strong with respect to the second.

另一方面，在诸如colonnade和kangaroo之类的单词中，双脚中的第二个更强，因为它有一个次要重读音节，然后是一个非重读音节，然后是一个主要重读音节：

In words such as colonnade and kangaroo, on the other hand, it is the second of the two feet which is the stronger, since it has a secondary stressed syllable followed by an unstressed syllable followed by a primary stressed syllable:

(5)

像champagne这样有一个主重读音节和一个副重读音节，但没有非重读音节的词包含两个脚，每个脚只包含一个强音节。但是，其中一只脚相对于另一只脚来说很强壮，因此：

Words such as champagne, which have a primary stressed and a secondary stressed syllable, but no unstressed syllables, contain two feet, each of which contains only a strong syllable. However, one of those feet is strong with respect to the other, thus:

(6)

在这个词中，两只脚中的第二只脚更强，而在名词export等词中，两只脚中的第一只脚更强：

In this word, the second of the two feet is the stronger, whereas, in a word such as the noun export, it is the first of the two feet which is the stronger:

(7)

回想一下，单音节虚词通常是非重读的（因此简单地用 W 标记）。词汇类别的单音节词可以与这些词形成分支脚，如短语hit it。因此，诸如hit it 之类的短语包含与witty等单个单词完全相同的脚部结构：

Recall that monosyllabic function words are typically unstressed (and are thus simply labelled with a W). Monosyllabic words of a lexical category may form branching feet with such words, as in the phrase hit it. Thus a phrase such as hit it contains exactly the same foot structure as a single word such as witty:

(8)

因此，我们称为脚的成分并不直接映射到单词上：一个单词中可能不止一只脚，一只脚可能超出单个单词的范围。此外，一个词可能无法完全分为英尺。例如，像America这样的词包含一个 foot，它由重读的 antepunultimate 音节和后面的两个非重读音节组成；词首的非重读音节是一个“杂散的”非重读音节，它是单词的一部分，但没有整合到由它后面的三个音节形成的脚部结构中（与 /s/ 前面的 /s/ 相同）一个起始辅音可能是一个词的一部分，但没有被整合到音节结构中）：

Thus, the constituent we have called the foot does not map directly onto the word: there may be more than one foot within a word, and a foot may extend beyond the span of a single word. Furthermore, a word may not be exhaustively divisible into feet. For example, words such as America contain a foot consisting of the stressed antepunultimate syllable and the two unstressed syllables which follow it; the word-initial, unstressed syllable is a ‘stray’ unstressed syllable, which is part of the word, but is not integrated into the foot structure formed by the three syllables which follow it (rather in the same way that an /s/ preceding an onset consonant may be part of a word without being integrated into syllable structure):

(9)

这里的词首，'stray'（非韵律）非重读音节与上面讨论的单音节功能词平行：在词的层面上，它没有整合到脚结构中。在这种分析英语脚结构的方法中，以非重读音节开头的单词，如America、about和maroon，并不由一个以 W 音节开头的脚组成，因为我们否认英语中有 WS 脚. 只有在较大的单位（例如短语）的水平上，此类非重读音节才可能被整合到脚部结构中，⁶如动词短语saw America中的那样：

The word-initial, ‘stray’ (extrametrical) unstressed syllable here is parallel to the monosyllabic function words discussed above: at the level of the word, it is not integrated into foot structure. On this way of analysing English foot structure, words which begin with an unstressed syllable, such as America, about and maroon, do not consist of a single foot which begins with a W syllable, since we are denying that there are W-S feet in English. It is only at the level of larger units such as the phrase that such unstressed syllables may be integrated into foot structure,6 as in the verb phrase saw America:

(10)

我们看到像champagne和export这样的词是双音节的并且包含两只脚。这些词与诸如maroon之类的词根本不同，后者也是双音节的，但以“杂散”非重读音节开头，并且仅包含一只脚，由重读音节组成：

We saw that words like champagne and export are bisyllabic and contain two feet. These are fundamentally different from words such as maroon, which are also bisyllabic but beging with a ‘stray’ unstressed syllable and contain only one foot, which consists of the stressed syllable:

(11)

9.3 音位概括和足部结构

9.3 Phonological Generalizations and Foot Stucture

将脚假设为语音成分的原因之一是，正如一些语音概括对音节结构敏感一样，一些语音概括也对脚结构敏感。以美国英语的许多方言中 Flapping 的概括或规则为例。在这种概括下，/t/ 和 /d/ 被实现为元音之间的肺泡拍打（也称为瓣膜），如Betty和bedding（[bεɾi]] 和 [bεɾiŋ]）。但如果脚边界出现在 /t/ 或 /d/ 附近，则该规则不适用。因此，概括不包括诸如attacker或a tacker 之类的情况，因为在这些情况下，脚边界介于第一个元音和 /t/ 之间，因此：

One of the reasons for postulating the foot as a phonological constituent is that, just as some phonological generalizations are sensitive to syllable structure, so some phonological generalizations are sensistive to foot structure. Take the generalization, or rule, of Flapping, in many dialects of American English. Under this generalization, /t/ and /d/ are realized as an alveolar tap (also known as a flap) between vowels, as in Betty and bedding ([bεɾi]] and [bεɾiŋ]). But the rule does not apply if a foot boundary occurs adjacent to the /t/ or /d/. Thus, the generalization does not cover cases such as attacker, or a tacker, since, in those cases, a foot boundary intervenes between the first vowel and the /t/, thus:

(12)

如前所述，如果一只脚由重读元音和紧随其后的任何非重读音节组成，那么像attacker这样的词包含一只脚（以重读音节开头），前面是一个“杂散”非重读音节，如在美国的词首音节中。因此词首音节不是 /t/ 出现的脚部的一部分，而在诸如Betty这样的词中，它是：

If a foot consists, as we have said, in a stressed vowel followed by any immediately following unstressed syllables, then a word such as attacker contains a single foot (which begins with the stressed syllable) preceded by a ‘stray’ unstressed syllable, as in the word-initial syllable of America. Thus the word-initial syllable is not a part of the foot in which the /t/ appears, whereas in a word such as Betty, it is:

(13)

请注意，拍打也发生在跨越单词边界形成的脚中，如hit it ([hɪɾɪt])：

Note that Flapping also occurs in feet which are formed across word boundaries, as in hit it ([hɪɾɪt]):

(14)

也就是说，Flapping 对单词边界或形态结构不敏感；相反，足部结构在拍打的应用中很重要：拍打仅适用于足部内部。

That is, Flapping is not sensistive to word boundaries or to the morphological structure; rather, it is foot structure which matters in the application of Flapping: Flapping only applies foot-internally.

通常被认为对足部结构敏感的概括的另一个例子是 Aspiration 规则。我们已经注意到，我们必须承认英语中清塞音的送气程度不同。然而，当所讨论的清音塞处于脚初始位置时，吸气最强，如在party和appearance 中，其脚结构在 (15) 中给出：

Another example of a generalization which is often said to be sensitive to foot structure is the rule of Aspiration. We have already noted that we must acknowledge that there are degrees of aspiration of voiceless stops in English. However, aspiration is at its strongest when the voiceless stop in question is in foot-initial position, as in party and appearance, whose foot structures are given in (15):

(15)

吸气最初适用于脚（尽管在其他位置可能会有某种程度的吸气）。

Aspiration applies foot-initially (although there may be some degree of aspiration in other positions).

9.4 英语节奏重现：重音节拍和韵律

9.4 The Rhythm of English Again: Stress Timing and Eurhythmy

我们在第 8 章中看到，英语的节奏是重读的。这意味着在说英语的人的演讲中发现的有规律的重复节拍（英语演讲的节奏）落在重读的音节上。也就是说，英语中的重读音节或多或少地以相等的间隔出现。在这方面，像英语这样的语言通常被认为与像法语这样的语言不同，因为在像法语这样的语言中，每个音节被认为以或多或少相等的间隔出现（因此，这类语言通常被称为音节计时）。

We saw, in chapter 8, that the rhythm of English is stress-timed. What this means is that the regular recurring beats found in the speech of English speakers (the rhythm of English speech) fall on stressed syllables. That is, stressed syllables in English occur at more or less equal intervals. Languages like English are often said to be distinct from languages like French in this respect in that, in languages like French, each syllable is said to occur at a more or less equal interval (languages of that sort are often, therefore, said to be syllable-timed).

这种节奏的后果之一是英语脚可能由一个重读音节和随后的一系列非重读音节组成，就像在公园里听到的短语中，其中重读音节后面跟着两个非重读音节，或短语heard it in the park，其中heard后面跟着 3，或者短语heard it in the announcement，后面跟着 4。

One of the consequences of this kind of rhythm is that English feet may consist of a stressed syllable followed by a sequence of unstressed syllables, as in the phrase heard in the park, in which the stressed syllable in heard is followed by two unstressed syllables, or the phrase heard it in the park, where heard is followed by three, or the phrase heard it in the announcement, where it is followed by four.

已经说过英语允许非常广泛的非重读音节序列，必须说“理想”或最佳节奏结构是强音节和弱音节以 SWSW 模式交替出现的结构。从感知的角度来看，这种“对立面交替”的序列似乎是最佳的：它们似乎使语音信号更容易解码。这种最佳的节奏结构通常被称为eurythmic结构。由此可知，最佳的、最有韵律的脚部结构是一种简单的 SW 结构，重读音节的右侧只有一个非重读音节。因此，与只有一个 W 音节的足部结构相比，具有多个 W 音节的足部结构更不和谐，更不理想，并且非重读音节的数量越多，脚部的节奏就越不和谐或最佳。

Having said that English allows for really quite extensive sequences of unstressed syllables, it has to be said that the ‘ideal’ or optimal rhythmic structure is one in which strong and weak syllables alternate, in an S-W-S-W pattern. It appears to be the case that such sequences of ‘alternating opposites’ are optimal in a perceptual sense: they seem to make the speech signal more easily decoded. Such optimal rhythmic structures are often referred to as eurhythmic stuctures. It follows from this that the optimal, most eurhythmic, foot structure is a simple S-W structure, with only one unstressed syllable to the right of the stressed syllable. Foot structures with more than one W syllable are therefore less eurhythmic, less optimal, than those with only one, and the greater the number of unstressed syllables, the less eurhythmic or optimal the foot.

这种对节奏的偏好延伸到脚的顺序：S 和 W 脚的顺序也比其他顺序更符合节奏。例如，在I want a cup of coffee这句话中，动词短语中有一个三英尺的SWS序列，每一个本身就是一个SW序列的音节；它在音节序列的水平和脚的序列水平上都是有韵律的：

This preference for eurhythmy extends to sequences of feet: sequences of S and W feet are also more eurhythmic than other sequences. For instance, in the sentence I want a cup of coffee, there is an S-W-S sequence of three feet in the verb phrase, each of which is itself an S-W sequence of syllables; it is eurhythmic both at the level of sequences of syllables and at the level of sequences of feet:

(16)

然而，在许多情况下，给定的单词组合可能会创建一个不合节奏的短语，并且实际上可能会导致相邻的 S 标记脚。这是因为在大多数英语短语中，最重读的是最后一个单词，例如前面讨论的短语black bird 。这种短语重音规则似乎适用于大多数类型的英语短语，如slowly ate（动词短语）、very yellow（形容词短语）、into London（介词短语）和very slowly（副词短语）。它似乎也适用于句子层面，正如我们从刚刚给出的例子中看到的那样：谓语动词短语比前面的主语名词短语更显着。在短语重音规则带来相邻 S 标记脚的地方，似乎可以采取“规避行动”。让我们考虑一些例子。

In many cases, however, a given combination of words may potentially create a phrase which is less than eurhythmic, and indeed may potentially result in adjacent S-labelled feet. This results from the fact that, in most English phrases, it is the final word which is most stressed, as in the phrase black bird, discussed earlier. This Phrasal Stress Rule seems to hold for most types of phrase in English, as in slowly ate (verb phrase), very yellow (adjective phrase), into London (prepositional phrase) and very slowly (adverb phrase). It also seems to apply at the level of the sentence, as we can see from the example just given: the predicate verb phrase is more salient than the preceding subject noun phrase. Where the Phrasal Stress Rule brings about adjacent S-labelled feet, it appears that ‘evasive action’ can be taken. Let us consider some examples.

以学术、田纳西和香槟这些词为例。很明显，academic重读倒数第二个音节，副重读第一个音节；其他音节不重读。字脚结构如下：

Take the words academic, Tennessee and champagne. Clearly, academic has primary stress on the penultimate syllable and secondary stress on the first syllable; the other syllables are unstressed. The foot structure of the word is as follows:

(17)

田纳西州也有两只脚，第二只比第一只强。然而，第二只脚只包含一个重读音节，后面没有非重读音节：

Tennessee also contains two feet, the second stronger than the first. However, the second foot consists simply of a stressed syllable, with no unstressed syllables following it:

(18)

香槟也有两个脚，正如我们已经看到的，第一个脚由一个带有次重音的音节组成，第二个脚由一个带有主重音的音节组成：

Champagne also has two feet, as we have already seen, the first of which consists of a syllable with secondary stress and the second of which consists of a syllable with primary stress:

(19)

在这三种情况中的每一种情况下，这个词都由两只脚组成，第二只脚比第一只脚强。然而，当这些词出现在两只脚中较重的脚后跟另一只脚的重读音节的短语中时，并且该音节必须比前一个更重读，就会产生一种“重音冲突”，其中，不是 S 和 W 脚的节奏序列，而是 SS 脚序列。在这种情况下，节奏反转规则适用。考虑一些这样的短语，例如学术玩笑、香槟早餐、Tennessee Williams. 请注意，在每种情况下，短语重音分配规则意味着两个词中的第二个词必须比第一个词具有更大的重音。另请注意，academic、champagne和Tennessee等词中的主要和次要重音已经颠倒了。也就是说，违规结构（在下面的 (20) 中举例说明）被更改为在 (21) 中举例说明的更加悦耳的结构。

In each of these three cases, the word consists of two feet, the second of which is strong with respect to the first. However, when these words appear in phrases where the stronger of the two feet is immediately followed by the stressed syllable of another foot, and where that syllable must be more heavily stressed than the preceding one, a kind of ‘stress clash’ results, in which, rather than a eurhythmic sequence of S and W feet, an S-S sequence of feet occurs. In situations such as this, a rule of rhythm reversal applies. Consider some such phrases, e.g. academic banter, champagne breakfast, Tennessee Williams. Note that, in each case, the rule for phrasal stress assignment means that the second of the two words must have greater stress than the first. Note too that the primary and secondary stresses in the words academic, champagne and Tennessee have reversed. That is, the offending structure (exemplified in (20) below) is altered to the more eurhythmic structure exemplified in (21).

(20)

(21)

这种节奏反转的过程在英语中是很有规律的。其他例子很容易找到；考虑皮卡迪利vs皮卡迪利圆环，希思罗vs希思罗机场，邓迪vs邓迪果酱，三十四本书vs三十四本书，好看的导师vs好看的导师，等等。正如我们所见，在英语短语中，中心词而不是前面的修饰语承受了最大的重音。每当包含脚的弱-强序列的单词与第一个音节是脚的第一个音节（即重读）的单词组合形成短语或复合词时，就会发生节奏反转。也就是说，在短语和复合词的上下文中，节奏反转在脚上而不是音节上起作用，反转脚的弱 - 强序列，而不是音节的弱 - 强序列。另一种表达方式是，当短语中的主重读音节后跟一个主重读音节时，反转过程会反转一个次要重读音节和一个主重读音节的序列。

This process of rhythm reversal is quite regular in English. Other examples are easily found; consider Piccadilly vs Piccadilly Circus, Heathrow vs Heathrow Airport, Dundee vs Dundee marmalade, thirty four vs thirty four books, good-looking vs good-looking tutor, and so on. As we have seen, in English phrases, it is the head, rather than a preceding modifier, which bears the most stress. Rhythm reversal occurs whenever a word containing a weak–strong sequence of feet is combined, to form a phrase or compound, with a word whose first syllable is the first syllable of a foot (i.e. is stressed). That is, rhythm reversal operates, within the context of phrases and compounds, on feet, not syllables, reversing weak–strong sequences of feet, rather than weak–strong sequences of syllables. Another way of putting this is to say that the reversal process reverses a sequence of a secondary stressed syllable and a primary stressed syllable when it is followed by a primary stressed syllable within a phrase.

反转不对由非重读音节和脚的第一个音节组成的序列进行操作，如在maroon sweater中。maroon一词包含一个单脚，仅由一个重读音节组成，没有后续的非重读音节；那只脚之前是一个“杂散”的非重读音节（就像美国的非重读音节一样，如上所示），因此：

Reversal does not operate on a sequence consisting of an unstressed syllable and the first syllable of a foot, as in maroon sweater. The word maroon contains a single foot, which consists only of a stressed syllable with no following unstressed syllables; that foot is preceded by a ‘stray’ unstressed syllable (just like the unstressed syllable in America, shown above), thus:

(22)

虽然栗色毛衣包含一个 SS音节序列，但它不包含一个 SS脚序列，因此它不会发生节奏反转。节奏逆转规则的普遍影响据说构成了证据脚的存在是英语音韵学中的一个组成部分。此外，像maroon这样的词没有发生逆转这一事实可以作为我们主张的证据 (a) 英语脚总是以重读音节开头，以及 (b) 英语单词不一定可以完全分为脚。换句话说，像maroon这样的词不能被分析为由具有 WS 音节序列的脚组成。

While maroon sweater contains an S-S sequence of syllables, it does not contain an S-S sequence of feet, and it does not therefore undergo rhythm reversal. The pervasive effects of the Rhythm Reversal Rule are said to constitute evidence for the existence of the foot as a consistuent in the phonology of English. Furthermore, the fact that words like maroon do not undergo reversal can be taken as evidence for our claims (a) that English feet always begin with a stressed syllable, and (b) that English words are not necessarily exhaustively divisible into feet. In other words, a word like maroon is not to be analysed as consisting of a foot with a W-S sequence of syllables.

节奏反转在脚序列水平而不是在音节序列水平上运行的说法得到以下事实的支持：它在诸如good-looking tutor之类的短语中运行，在反转之前，具有以下内容结构体：

The claim that rhythm reversal operates at the level of sequences of feet, rather than at the level of sequences of syllables, is supported by the fact that it operates in phrases such as good-looking tutor, which, prior to reversal, has the following structure:

(23)

如果反转在 S 和 W 音节序列（而不是脚）的水平上进行，它不会影响序列looking和tutor，它们具有交替的 S-WS-W 结构。违反 SW 排序的是脚的水平。

If reversal operated at the level of sequences of S and W syllables (rather than feet), it would not affect the sequence looking and tutor, which have an alternating S-WS-W structure. It is at the level of the foot that the S-W sequencing is violated.

颠倒最引人注目的方面是它展示了句法和音系的相互作用。反转操作的条件部分取决于关于英语的句法事实：修饰语通常位于英语短语中的中心词之前。这一点，再加上头部比修饰符承受更多压力的事实，导致了反转现象。

The most striking aspect of reversal is that it demonstrates the interaction of syntax and phonology. The conditions under which reversal operates are partly determined by a syntactic fact about English: the fact that modifiers typically precede heads in English phrases. This, combined with the fact that it is the head which receives more stress than the modifier, brings about the reversal phenomenon.

反转也与形态结构相互作用：它在包含本身带重音的后缀的单词中运行。以纽约这个词为例。它由两只脚组成，第二只脚比第一只脚强：

Reversal also interacts with morphological structure: it operates within words which contain a suffix which itself takes stress. Take the word New York. It consists of two feet, the second stronger than the first:

(24)

后缀-ese是那些带有重音的英语后缀之一。它由一只不分支的脚组成，当它被添加到New York时，生成的单词New Yorkesese由 WSS 序列中的三只脚组成：

The suffix -ese is one of those English suffixes which takes stress. It consists of a single non-branching foot, and when it is added to New York, the resulting word New Yorkese consists of three feet in a W-S-S sequence:

(25)

此结构满足反转条件，然后应用于产生纽约语中的 SWS 脚序列。

This structure meets the conditions for reversal, which then applies to yield the S-W-S sequence of feet in New Yorkese.

当然，事实上，人们可以说出诸如学术玩笑之类的短语，主要重音放在学术的倒数第二个音节上。当一个人这样做时，通常是将短语的某些方面与其他可能性进行对比。例如，有人可能会强调，one 指的是学术玩笑，而不是其他类型的玩笑，例如青少年玩笑。这种现象最重要的事实是它是一种话语现象，通常被称为对比句重音或对比语调（详见第 10 章）：如果没有适当的话语设置，一个人就无法理解这样的强调，在这个话语设置中，一个人的对话者有机会理解一个人正在对比节奏反转形容词学术的其他可能性。下面的交换是一个例子：

It is true, of course, that one can, in fact, utter phrases such as academic banter with primary stress on the penultimate syllable of academic. When one does this, one is usually contrasting some aspect of the phrase with some other possibility. One might be stressing, for instance, that one means academic banter and not some other kind of banter, such as adolescent banter. The most important fact about this phenomenon, often referred to as contrastive sentence stress or contrastive intonation (see chapter 10 for details), is that it is a discourse phenomenon: one could not make sense of such a stressing if it were not accompanied by an appropriate discourse setting in which one’s interlocutor stands a chance of understanding what other possibilities one is contrasting the rhythm-reversed adjective academic with. The following exchange is an example:

(26)	A：	我确实喜欢学术玩笑，你知道的。（节奏反转）
	乙：	开什么玩笑？
	A：	ˌ学术玩笑。（对比压力；无节奏逆转）

尽管如此，这里的重音模式与上面 (20) 中给出的重音模式不同，因为banter的重音音节比academic的主要重音音节重音小。⁷这些例子表明，反转是一种与形态和句法相互作用的格律现象，可以独立于语篇语境来描述，而对比语调是一种不能独立于语篇语境来描述的现象。这表明，区分那些可以独立于话语语境进行分析的现象与那些不能独立分析的现象是可能的，而且也许是必要的。

The stress pattern here is nonetheless distinct from that given in (20) above, since the stressed syllable of banter is less stressed than the primary stressed syllable of academic.7 What examples such as these suggest is that reversal is a metrical phenomenon which interacts with morphology and syntax, and can be described independently of discourse context, whereas contrastive intonation is a phenomenon which cannot be described independently of discourse context. This suggests that it is possible, and perhaps necessary, to distinguish those phenomena which can be analysed independently of context of utterance from those which cannot.

笔记

Notes

1也有不完全押韵，其中重读音节后的语音片段相似但不完全相同。单词manner ([ˈmænə]) 和banger ([ˈbæŋə]) 之间的不完全押韵就是这样的例子：鼻音的发音位置不一样，所以banner与manner押韵更好。但这毕竟不是那么糟糕的押韵：[n] 和 [ŋ] 都是鼻音，所以它们听起来非常相似。

1 There are also imperfect rhymes, where the phonetic segments following the stressed syllable are similar but not identical. The imperfect rhyme between the words manner ([ˈmænə]) and banger ([ˈbæŋə]) are examples of this: the place of articulation of the nasals is not the same, so banner would rhyme better with manner. But it’s not that bad a rhyme, after all: both [n] and [ŋ] are nasals, so they sound very similar.

2这里还有其他因素在起作用，包括语素flexi的可用性（如flexitime，它本身是通过类比加班形成的；flexi本身来自形容词flexible e，其中前缀是flex）。术语pescitarian取决于拉丁词“鱼”的知识可用性。

2 There are other factors at play here, including the availability of the morpheme flexi (as in flexitime, itself formed by analogy with overtime; flexi itself comes from the adjective flexible, in which the prefix is flex). The term pescitarian depends on the availability of knowledge of the Latin word for ‘fish’.

3在这种情况下，形态和韵律结构是一致的。

3 In this case, the morphology and the metrical structure coincide.

4由于篇幅所限，我们在此展示的格律树已被缩写。由于韵律结构是由音节结构决定的，严格来说，我们应该展示建立在音节结构树上的韵律树，如下所示：

4 The metrical trees we present here are abbreviated, for reasons of lack of space. Because metrical structure is determined by syllable structure, we ought, strictly speaking, to show metrical trees built upon syllable structure trees, as follows:

本章中使用的缩写树实际上并没有表明是韵律的结构，而不是整个音节，这对决定韵律结构至关重要。但是，它们足以满足我们的目的。

The abbreviated trees used in this chapter do not actually show that it is the structure of the rhyme, and not the entire syllable, which is crucial in determining metrical structure. However, they will suffice for our purposes.

本章中使用的缩写树实际上并没有表明是韵律的结构，而不是整个音节，这对决定韵律结构至关重要。但是，它们足以满足我们的目的。

5我们用“S”标记非分支脚，这与我们对词汇类别的单音节词的处理一致。

5 We label the non-branching foot with an ‘S’, consistent with our treatment of monosyllabic words of a lexical category.

6英语中脚结构和词结构不对齐的可能性经常被词曲作者利用，他们根据句法（词和短语）、节奏（韵律结构）和各种重复类型（例如头韵）来构建歌曲, 韵律和音位结构的押韵和重复。正是出于这个原因，可以构建涉及礼仪([ˈmænəz])、语法([ˈgɹæmə]) 和单词damn的重复，然后是第一个杂散的、非重读的音节about（产生一只脚，其形式为[dæmə]）。在这里，重复利用音节和脚结构，同时跨越单词和短语结构。因此，这种被称为“押韵”的重复可以延伸到我们称为押韵的音节成分之外，进入脚部结构。歌曲“钻石是女孩最好的朋友”中continental和rental的押韵就是一个例子。

6 The possibility of non-alignment of foot structure and word structure in English is often exploited by songwriters, who structure their songs on the basis of syntax (words and phrases), rhythm (metrical structure) and various types of repetition, such as alliteration, rhyming and repetition of metrical and phonemic structures. It is for this reason that it is possible to construct a repetition involving manners ([ˈmænəz]), grammar ([ˈgɹæmə]), and the word damn followed by the first, stray, unstressed syllable of about (yielding a foot whose form is [ˈdæmə]). Here, a repetition exploits syllable and foot structure while cutting across word and phrase structure. The sort of repetition known as ‘rhyming’ can therefore extend beyond the syllabic constituent we have called the rhyme, into foot structure. The rhyming of continental and rental in the song ‘Diamonds are a girl’s best friend’ is a further example.

7通常认为，这是因为“玩笑”在话语语境中以某种方式“被给予”。

7 It is often suggested that this follows from the fact that ‘banter’ is somehow ‘given’ in the discourse context.

练习

Exercises

1 为下列每个单词画出格律树。首先绘制足部结构。如果一个词包含不止一只脚，画出一个上级的 SW 或 WS 分支结构，显示两只脚中哪只脚更强。

（一个漂亮

(b) 相撞

(d) 蝙蝠

(e) 夜莺

(f) 袋鼠

(g) 拉比

(h) 合同

gueqrecard_001

2 收听www.wiley.com/go/carrphonetics上的曲目 9.1，并为录音中听到的以下短语绘制韵律树：

(a) 解雇一名工人

(b) 高兴的代理人

(d) Piccadilly Circus（在应用节奏反转之前和之后显示最后一个短语的树）

3 听音轨 9.2并解释录音中所听到的以下表达的韵律结构：

(a) 不明智的决定

(b) 一间装修精美的卧室

(d) 一个心碎的人

(e) 十五美元

(f) 五十磅

(g) 深绿色长裤

(h) 香槟鸡尾酒

gueqrecard_001

4 进一步的音标练习

以尽可能多的语音细节转录以下在Track 9.3中说出的单词，指出音节边界、主要重音和（如适用）次要重音：

(a) 基本的

(b) 不友善

(d) 机会

(e) 中文

10 英语语调

English Intonation

10.1 主音音节、声调和语调短语

10.1 Tonic Syllables, Tones and Intonation Phrases

虽然我们经常说某些说话者说话方式单调，但事实是人类不会发出本质上单调的语音：我们都会改变我们的语音，创造语调轮廓。但究竟什么是语调？它是在话语中使用音调变化。什么是音高？我们已经知道，音高是由声带振动速率的变化产生的听觉印象。语调是在通常由多个单词组成的一段语音上使用音调轮廓。一个例子是话语Mary went to the doctor。这段话中有三个主重音音节：倒数第二个音节Mary、单音节went和倒数第二个音节doctor. 但是在主重读音节doctor上有额外的音高移动。该重读音节在感知上比其他重读音节更突出，并且往往比话语中的其他重读音节持续时间更长，声音更大。那个音节据说是补品音节。“主音”这个词表示这个音节是声调落下的地方。音调是放置在该音节上的额外音高运动。在我们的示例中，音调是降调：随着音节的发出，声带的振动速率降低，导致音调从高音过渡到低音。我们将这些表示如下：

Although we often say that some speakers speak in a monotonous manner, the fact is that human beings do not utter speech which is monotone in nature: we all inflect our speech, creating intonational contours. But what is intonation, exactly? It is the use of pitch variation in discourse. What is pitch? We have seen that pitch is the auditory impression created by variations in the rate of vibration of the vocal folds. Intonation is the use of pitch contours over stretches of speech which often consist of more than one word. An example is the utterance Mary went to the doctor. There are three syllables with primary word stress in this utterance: the penultimate syllable of Mary, the single syllable of went and the penultimate syllable of doctor. But there is additional pitch movement on the primary-stressed syllable of doctor. That stressed syllable is perceptually more prominent than the others, and will tend to be longer in duration, and louder, than the other stressed syllables in the utterance. That syllable is said to be the tonic syllable. The word ‘tonic’ denotes the fact that this syllable is where the tone falls. The tone is the extra pitch movement placed on that syllable. In our example, the tone is a falling tone: the rate of vibration of the vocal folds decreases as the syllable is uttered, resulting in a transition from a higher to a lower pitch. We will represent these as follows:

gueqrecard_001

(1) Mary 去看了↘医生。（在www.wiley.com/go/carrphonetics上跟踪 10.1）

正如我们在第 8 章的单词重读中看到的，“Mary”和“went”上的变音符号表示以下音节被重读。↘ doctor上的下划线表示它是主音节（因此，根据定义，强调），前面的 '↘' 变音符号表示降调。这种语气是典型的陈述性话语，说话者在做陈述，而不是提出是/否的问题。其他音调在英语中也是可能的。在是/否问题（可能征求回答“是”或“否”的问题）中，常见的是在主音节中发现升调，例如在问题“玛丽怀孕了吗？”中我们将表示升调如下：

As we saw in chapter 8 on word stress, the diacritics on ˈMary and ˈwent indicate that the following syllables are stressed. The underlining on ↘doctor indicates that it is the tonic syllable (and thus, by definition, stressed), and the preceding ‘↘’ diacritic indicates a falling tone. This kind of tone is typical of declarative utterances, in which the speaker is making a statement, as opposed to, say, posing a yes/no question. Other tones are possible in English. In yes/no questions (questions which may solicit the responses ‘Yes’ or ‘No’), it is common to find a rising tone in the tonic syllable, as in the question Is Mary pregnant? We will represent rising tones as follows:

gueqrecard_002

(2) Mary ↗怀孕了吗？（曲目 10.2）

这里倒数第二个音节Mary和Pregnant重读，主音落在Pregnant的重音节上，声调是升调。

Here, the penultimate syllables of Mary and pregnant are stressed, the tonic falls on the stressed syllable of pregnant, and the tone is a rise.

第三种音调是升降调，音高先升后降，如下面的对话：

A third tone is the rise-fall tone, in which the pitch rises and then falls, as in the following exchange:

(3)	妻子：	你见过↗玛丽吗？
	丈夫：	↗↘不！（曲目 10.3）

上升下降声调的使用传达了说话者的确定性、感叹、强烈的信念或强烈的感情。在这种情况下，丈夫是在说他肯定没有见过玛丽：语调传达出对暗示指控的完全否认。

The use of rise-fall tones conveys certainty, exclamation, strong conviction or strength of feeling on the part of the speaker. In this case, the husband is saying that he has certainly not been seeing Mary: the intonation conveys a complete denial of the implied accusation.

第四声是降升调，如下面的交流：

A fourth tone is the fall-rise tone, as in the following exchange:

gueqrecard_002

(4)	妻子：	你见过↗玛丽吗？
	丈夫：	↗↘不！（曲目 10.4）

在这里，在第二个话语中，音高先降后升。使用这种语气表示说话者犹豫、不确定、搪塞或保留。在这个例子中，丈夫的回答不够明确和直截了当：他否认他一直在和玛丽约会，或者试图暗示他一直在做的事情并不等于浪漫意义上的“看到玛丽”。

Here, the pitch falls then rises in the second utterance. Use of such a tone conveys hesitation, lack of certainty, prevarication or reservation on the part of the speaker. In this example, the husband is being less than clear and straightforward in his response: he is denying that he’s been seeing Mary, or trying to suggest that what he’s been doing does not really amount to ‘seeing Mary’ in the romantic sense.

包含主音节的一段语篇称为语调短语(IP)，也称为语调群、语调单元或声调群。这些也被称为呼吸群，因为它们构成了我们从肺部排出空气的语言单位。当我们停止说话以换气时，我们通常会在音调组结束时这样做：说话者通常会在此类单元结束时停顿。

A stretch of discourse which contains a tonic syllable is called an intonation phrase (IP), otherwise known as an intonation group, intonation unit or tone group. These are also referred to as breath groups, since they constitute units of speech in which we expel air from the lungs. When we stop speaking to draw breath, we often do so at the end of a tone group: it is common for speakers to pause at the end of such units.

通常识别语调的三个主要特征：

It is common to identify three main features of intonation:

• 将一段语音分成 IP，

• 将主音放在该块的重读音节之一上，并且

• 指定主音音节上的特定音调

在某种程度上，这些是自变量：我们选择将主音放在哪个音节上可以独立于 IP 边界的位置，我们在主音音节上放置的声调可以独立于我们选择放置主音的位置。

These are, to some extent, independent variables: what syllable we choose to put the tonic on can be independent of where the IP boundaries go, and what tone we place on the tonic syllable can be independent of where we choose to place the tonic.

在示例 (1) 到 (4) 中，主音落在所谓的最后一个词项(LLI) 上。回想一下我们对单词重音的讨论，单词可以分为两大类：词汇类别的单词（通常是名词、动词、形容词和副词）和功能或语法类别的单词（例如冠词、连词、介词和代词）。因此，句法单元中的最后一个词项是最后一个名词、动词、形容词或副词。例如，在 (1) 中，LLI 是名词doctor，而在 (2) 中，它是形容词pregnant。以下示例在 (5) 中包含动词 LLI，在 (6) 中包含副词 LLI：

In examples (1) to (4), the tonic falls on what is known as the last lexical item (LLI). Recall from our discussion of word stress that words can be classified into two broad groupings: words of a lexical category (typically nouns, verbs, adjectives and adverbs) and words of a functional, or grammatical, category (such as articles, conjunctions, prepositions and pronouns). The last lexical item in a syntactic unit is thus the last noun, verb, adjective or adverb. For example, in (1), the LLI is the noun doctor, and in (2) it is the adjective pregnant. The following examples contain, in (5), an LLI which is a verb, and in (6), an LLI which is an adverb:

gueqrecard_002

(5) 我老公↘出轨。（曲目 10.5）

(6) His ˈlover ˈwalks ↘ grace fully. （曲目 10.6）

LLI 可能不是 IP 中的最后一项，如：

The LLI may not be the last item in an IP, as in:

(7) ˈBill ↘给了她。（曲目 10.7）

在这里，最后一项是代词，不是词项，因此不带主音。倒数第二项是介词，因此也不能服用补品。倒数第三项是代词，所以那太不能服用补品了。补品落在gave上，因为它是 LLI，但不是最后一项。在这种情况下，主音音节之后的任何音节都被认为构成了 IP 的尾部：在这里下降之后，音高只是在低水平上下降到主音之后的其余音节中。

Here, the last item is a pronoun, which is not a lexical item, and thus does not take the tonic. The second-last item is a preposition, and thus also fails to take the tonic. The third-last item is a pronoun, so that too fails to take the tonic. The tonic falls on gave, since it is the LLI, but not the last item. In cases like this, any syllables which follow the tonic syllable are said to constitute the tail of the IP: after the fall here, the pitch just trails off at a low level into the remaining syllables after the tonic.

10.2 偏离 LLI 规则

10.2 Departures from the LLI Rule

LLI 规则是放置主音的默认规则。“默认”是指在没有特殊情况下放置补品的位置。语言学中的默认值很像您计算机上的默认设置：除非有人为了某些特殊目的故意更改设置，否则它们是使用的设置。出于各种目的，将主音从默认位置移开在英语中很常见。我们现在检查其中的一些。

The LLI rule is the default rule for the placement of the tonic. By ‘default’, we mean the point where the tonic is placed if no special circumstances prevail. Defaults in linguistics are rather like the default settings on your computer: they are the settings that are used unless one deliberately changes the set-up for some special purpose. It is common in English to shift the tonic away from the default position, for various purposes. We now examine some of those.

10.2.1 对比语调

10.2.1 Contrastive Intonation

gueqrecard_002

(8)	演讲者甲：	ˈMary ˈ给了 ˈJohn 一个 ↘相机。
	演讲者乙：	不，她没有给他；他给了她。( Track 10.8 ) (= ↘没有 \| ↘她，没有，给 ↘他 \| ↘他，给了 ↘她)

在这里，斜体表示，除了No之外，还有四个代词接受了补品（垂直线表示 IP 之间的边界）：不是Mary给了John a camera：是John给了Mary a camera。一旦说话者 A 说话，这里给出相机一词的指称。考虑以下句子John is taking the train to London可能的语调模式：

Here, the italics show that, in addition to No, four pronouns receive the tonics (the vertical lines indicate the boundary between the IPs): it wasn’t Mary that gave John a camera: it was John that gave Mary a camera. The referent of the word camera here is given, once speaker A has spoken. Consider the following possible intonational patterns for the sentence John is taking the train to London:

gueqrecard_002

(9) ˈJohn is ˈtaking the ˈtrain to ↘ Lon don. （曲目 10.9）

(10) ˈJohn is ˈtaking the ↘ train to ˈLondon. （曲目 10.10）

(11) ↘约翰正在搭火车去伦敦。（曲目 10.11）

在 (9) 中，我们有主音放置的默认模式，主音位于 LLI 的重音音节上。在 (10) 中，我们有对比语调：将火车与其他一些交通工具（例如飞机）进行对比。在 (11) 中，说话者强调了一个事实，即乘火车去伦敦的是约翰，而不是其他人。

In (9), we have the default pattern for tonic placement, with the tonic on the stressed syllable of the LLI. In (10), we have contrastive intonation: the train is being contrasted with some other mode of transport, such as the plane. In (11), the speaker is stressing the fact that it is John, not someone else, who is taking the train to London.

这种补品放置的使用与所谓的焦点有关。在 (9) 中，我们有广泛的关注点，与一切都是新闻的陈述相关联。这些陈述被称为“突如其来”：所有信息都作为新信息宣布，因此话语中的所有内容都成为焦点。

This use of tonic placement relates to what is called focus. In (9), we have broad focus, associated with statements in which everything is news. These are statements which are said to come ‘out of the blue’: all of the information is announced as new information, so everything in the utterance is brought into focus.

在 (10) 中，被寻呼的人已经知道约翰要去伦敦：什么是新闻是关于他的交通方式的信息。这称为窄焦点。

In (10), the person being addressed already knows that John is going to London: what is news is the information concerning his mode of transport. This is called narrow focus.

在 (11) 中，收件人知道有人乘火车去伦敦，并被告知此人是约翰。这也是一个焦点狭窄的例子。狭隘的焦点与给定/新的区别有关。给定信息是共享（相互）知识，说话者和听者都知道。说话者和听者以前不知道新信息。

In (11), the addressee knows that someone is taking the train to London, and is being informed that the person in question is John. This too is a case of narrow focus. Narrow focus relates to the given/new distinction. Given information is shared (mutual) knowledge, known to both the speaker and the hearer. New information is not previously known to the speaker and the hearer.

为了对比的目的，包括词缀在内，主音几乎可以移到任何音节上。以下是英国首相戴维·卡梅伦 (David Cameron) 在 2010 年发表的声明：

The tonic can be moved onto almost any syllable for contrastive purposes, including affixes. Here is a statement by the British prime minister David Cameron in 2010:

(12)将要创造的不是 ↘ unˈployment 。| 这是 ↘ em ˈployment。

通常，employment和unemployment的主要重音在倒数第二个音节上，例如We're ˈtrying to creˈate em ↘ ploy ment。在这里，总理将就业与失业进行了对比。我们从这个例子中看到，在我们希望突出显示给定单词以便与另一个单词对比的上下文中，补品可能被放置在 LLI（第一个 IP 中的unemployment）之外的其他东西上，甚至附加可能会得到补品（失业中的非前缀）。

Normally, employment and unemployment have primary stress on the penultimate syllable, as in We’re ˈtrying to creˈate em↘ployment. Here, the prime minister is contrasting employment with unemployment. We see from this example that, given a context in which we wish to highlight a given word in order to contrast it with another, the tonic may be placed on something other than the LLI (unemployment in the first IP), and that even affixes may receive the tonic (the un- prefix in unemployment).

10.2.2 给定信息

10.2.2 Given Information

另一种违反 LLI 规则的情况涉及给定信息和新信息的概念。考虑以下交换：

Another situation in which the LLI rule is flouted concerns the notions of given and new information. Consider the following exchange:

gueqrecard_002

(13)	A：	我们需要西红柿。（我们需要↘ ma toes）
	乙：	我们有西红柿！（我们有 ↘得到了matoes）（曲目 10.12）

西红柿这个词在第一个陈述中给出：它已经被提及，所以它所传达的信息是给定的（交流参与者共享）。因此，补品从 LLI（西红柿）转移到了got一词上。现在考虑以下问题：

The word tomatoes is given in the first statement: it has already been mentioned, so the information it conveys is given (shared by the participants in the exchange). The tonic is therefore shifted away from the LLI (tomatoes) onto the word got. Now consider the following:

(14) 在↘大多数情况下，| 我们应用↘规则，| 但在 ↘某些情况下，| 我们 ↘不。（曲目 10.13）

在这里，第一个 IP（案例）中的 LLI 由话语上下文给出：如果我们说出 (14)，我们正在与之交谈的人已经知道规则是关于什么的，以及正在谈论什么类型的案例。

Here, the LLI in the first IP (cases) is given by the context of utterance: if we utter (14), the person we are speaking to already knows what the rule is about, and what kinds of cases are being spoken about.

同义词可以算作传达给定信息：

Synonyms can count as conveying given information:

(15)	A：	她“借”了简的 ↘连衣裙。
	乙：	↘否, \| 这是 ↘ Mary 's dress（曲目 10.14）

这里没有给出dress这个词，但它的意思是通过同义词frock的发音。

Here, the word dress isn’t given, but its meaning is, via the uttering of the synonym frock.

预设可以通过主音放置来传达：

Presuppositions can be conveyed via tonic placement:

(16)	A：	你和 ↗约翰说过话了吗？
	乙：	我不 ↘和种族主义者说话（曲目 10.15）

B 假设“约翰是种族主义者”已获得信息。所以主音从 LLI（种族主义）转移到前面的词项。说话者 A 可以强加约翰是种族主义者的假设，即使这是有待商榷的。

B is presupposing that ‘John is a racist’ is given information. So the tonic shifts from the LLI (racist) to the preceding lexical item. Speaker A can impose the presupposition that John is a racist, even if that is open to question.

注意这里动词speak是给定的，但仍然带主音：对比语调会导致主音被置于给定的信息上。这是另一个例子：

Notice here that the verb speak is given, but nonetheless takes the tonic: contrastive intonation can lead to the tonic being placed on given information. Here is another example:

gueqrecard_002

(17)	A：	他要去↘巴黎。
	乙：	他不会去↘巴黎。\| 他要去↘伦敦。（曲目 10.16）

Given information can be shared by millions of people (eg the fact that Barack Obama was elected president of the United States) or by as few as two people (eg husband and wife).

Given information can be shared by millions of people (e.g. the fact that Barack Obama was elected president of the United States) or by as few as two people (e.g. husband and wife).

10.2.3 词尾时间状语

10.2.3 Final Temporal Adverbials

通常会发现句法单元中具有状语功能并传达与时间有关的信息的 LLI 无法采用主音，例如：

It is common to find that LLIs which are in syntactic units which have an adverbial function, and which convey information relating to time, fail to take the tonic, as in:

gueqrecard_002

(18) ˈJohn's ˈgoing to ↘ Lon don on ˈSaturday. （曲目 10.17）

在这里，Saturday是 LLI，但是由于介词短语on Saturday是一个最后的时间状语，所以该状语表达式中的 LLI 没有采取补品。如果我们将补品放在星期六，那将构成对比语调的情况：

Here, Saturday is the LLI, but since the prepositional phrase on Saturday is a final temporal adverbial, the LLI within that adverbial expression fails to take the tonic. If we were to place the tonic on Saturday, that would constitute a case of contrastive intonation:

(19) ˈJohn's ˈgoing to ˈLondon on ↘ Saturday ( Track 10.18 )（相对于一周中的其他日子）。

当词尾状语放在前面时，它们往往会形成一个单独的 IP：

When final adverbial expressions are fronted, they tend to form a separate IP:

(20) 在 ↘星期二，| ˈJohn's ˈgoing to ↘伦敦。（曲目 10.19）

10.2.4 “事件”语句

10.2.4 ‘Event’ Sentences

这些比较好奇。它们是包含不及物动词的简短陈述，但主音没有落在不及物动词上而不是 LLI：

These are rather curious. They are short statements which contain intransitive verbs, but the tonic fails to fall on the intransitive verb rather than the LLI:

gueqrecard_002

(21)	（一种）	↘水壶在沸腾。
	（乙）	↘ ba by 的哭声。
	（C）	你的 ↘房子着火了。
	(四)	↘太阳出来了。（曲目 10.20）

人们会期望主音落在 LLI沸腾、哭泣和火上，以及助词out上（见上文 LLI，以及下文关于不及物短语动词，我们期望补音落在助词上）。

One would have expected the tonics to fall on the LLIs boiling, crying and fire, and on the particle out (see above on LLIs, and below on intransitive phrasal verbs, where we expect the tonic to fall on the particle).

据观察，这些句子中的主语并不表示人类代理人，但为什么这会影响补品的位置还远不清楚。在这种情况下，主题似乎有务实的前景化（选择出来）。

It has been observed that the subjects in such sentences do not denote human agents, but why that should affect the tonic placement is far from clear. There seems to be pragmatic foregrounding (selecting out) of the subject in such cases.

10.2.5 常带进补的非词项

10.2.5 Non-Lexical Items which Often Take the Tonic

非词汇项someone, something, somewhere, somebody ( no one, nothing, nowhere, nobody ) 的否定等价物经常采取补品：

The negative equivalents of the non-lexical items someone, something, somewhere, somebody (no one, nothing, nowhere, nobody) often take the tonic:

gueqrecard_002

(22)	（一种）	我看到 ↘没有人。
	（乙）	我什么都没做↘ 。
	（C）	我们没有得到 ↘在哪里。
	(四)	此 ˈinterests ↘无体。（曲目 10.21）

可以将主音放在非词汇项目someone, something, somewhere和someone上，但前提是语调是对比的，例如：

One can place the tonic on the non-lexical items someone, something, somewhere and somebody, but only if the intonation is contrastive, as in:

(23)	A：	今天早上我在↘松林里看到了邻居。
	乙：	你 ↘不可能有。\| 他现在在 ↘Paris。
	A：	好吧，我看到了 ↘一些。（轨道 10.22）（即，不是邻居，而是其他人）。

非母语人士应该知道，诸如one和do so 之类的前形式不是词汇项目：它们传达给定的信息，因此通常不带主音，如：

Non-native speakers should be aware that pro-forms such as one and do so are not lexical items: they convey given information, and thus do not normally take the tonic, as in:

(24)

10.2.6 分句

10.2.6 Cleft Sentences

分句采用以下形式：

Cleft sentences take the following form:

(25) 穿苏格兰短裙的是苏格兰人。

是比尔干的。

It was Bill that did it.

分裂是一种突出或突出句法成分的方式。可以说：

Clefting is a way of highlighting, or bringing into focus, a syntactic constituent. One could say:

(26) 我喜欢 ↘约翰。

但在裂口版本中，更强调了约翰与其他任何人之间的对比：

But in the cleft version, the contrast between John and anyone else is more emphasized:

gueqrecard_002

(27) It's ↘ John that I love. （曲目 10.24）

爱虽然是LLI，但补品落在了高亮项上。在这里，高亮材料之后的材料构成了IP的尾部：听者知道说话者爱某人，所以知识被给予，不需要进一步的补品。

Although love is the LLI, the tonic falls on the highlighted item. Here, the material after the highlighted material forms the tail of the IP: the hearer knows that the speaker loves someone, so that knowledge is given, and no further tonic is required.

10.2.7 指示性表达

10.2.7 Deictic Expressions

deictic一词的意思是“参与指出”，即在说话时用手指实际指向某物，或者在没有物理指向的情况下引起某人的注意。英语中的指示性表达包括指示词this、that、these和those。这些算作功能词，因此当 LLI 规则适用时，它们不会服用补品，如以下问题：

The word deictic means ‘involved in pointing out’, either by literally pointing at something with one’s finger while speaking, or bringing something to someone’s attention without physically pointing. Deictic expressions in English include the demonstrative words this, that, these and those. These count as function words, so they do not take the tonic when the LLI rule applies, as in the following questions:

gueqrecard_002

(28)	（一种）	你能↗给我吗？
	（乙）	我可以要 ↗五个吗？（曲目 10.25）

在这样的话语中，从话语的上下文中可以清楚地看出那个和那些被用来指代什么（在第一个例子中，父母可能要求孩子递给她一把刀；在第二个例子中，购物者可能是请店主给她五个橘子）。

In such utterances, it is clear from the context of utterance what that and those are being used to refer to (in the first example, a parent might be asking a child to hand her a knife; in the second example, a shopper might be asking a shopkeeper to give her five oranges).

如果明确提及所指事物，则 LLI 规则会将主音分配给所讨论的名词，如：

If the thing being referred to is explicitly mentioned, the LLI rule will assign the tonic to the noun in question, as in:

(29)	（一种）	你能给我那把↗刀吗？
	（乙）	我可以吃这些 ↗或anges 中的那些吗？（曲目 10.26）

当它们被对比使用时，补品可以落在指示性表达上，例如：

The tonic can fall on deictic expressions when they are being used contrastively, as in:

gueqrecard_002

(30)	（一种）	你能给我↗那把刀吗？（相对于其他刀）
	（乙）	我可以吃那些↗中的↗橙子吗？（轨道 10.27）（相对于其他一些橙子品种）

10.3 IP 和句法单元

10.3 IPs and Syntactic Units

10.3.1 通常构成独立 IP 的句法单元

10.3.1 Syntactic Units which Normally Form a Separate IP

有句法单元通常形成单独的 IP，例如相对较短的主句，如 10.1 节中的 (1) 至 (3)：这里，我们可以看到语调跟踪句法。这并不奇怪，因为从句和 IP 都传达了连贯的信息块。我们现在将检查通常形成独立 IP 的一系列其他句法单元。

There are syntactic units which normally form separate IPs, such as relatively short main clauses, as in (1) to (3) in section 10.1: here, we can see that the intonation tracks the syntax. This is unsurprising, since both clauses and IPs convey coherent chunks of information. We will now examine a range of other syntactic units which normally form independent IPs.

10.3.1.1 括号

10.3.1.1 Parentheticals

附加信息是演讲者提供的额外的可选信息。如果句法结构中省略了括号，则所讨论的结构在语法上仍然是合式的。让我们看看一些类型的括号。

Parenthetical information is extra, optional information offered by the speaker. If parentheticals are omitted from a syntactic structure, the structure in question remains grammatically well-formed. Let us look at some types of parenthetical.

非限制性关系从句

Non-restrictive relative clauses

gueqrecard_002

(31) ↘车里的那些家伙，| 谁↘饿了， | ˈ吃了一些↘三明治。（曲目 10.28）

此处，IP 边界对应于句子书面形式中的逗号。补品落在每个 IP 中的 LLI 上：汽车、饥饿和三明治。传达的意思是车上的人都饿了（因此表达“非限制性关系从句”：指代范围不受限制）。

Here, the IP boundaries correspond to the commas in the written form of the sentence. The tonics fall on the LLIs in each IP: car, hungry and sandwiches. The meaning conveyed is that all of the guys in the car were hungry (thus the expression ‘non-restrictive relative clause’: the range of referents is not restricted).

（限制性关系从句

(Restrictive relative clauses

请注意，这些通常不算作括号，因此通常不会形成单独的 IP，如：

Note that these do not normally count as parentheticals, and thus do not normally form a separate IP, as in:

(32) ↘饿了的车里的那些人| ˈ吃了一些↘三明治。（曲目 10.28）

在第二个示例中，只有两个 IP，而不是第一个示例中的三个。主语名词短语足够长，可以形成一个单独的 IP，但该名词短语中的限制性关系从句不会形成一个单独的 IP。(31) 和 (32) 的意义不同之处在于，在 (32) 中不一定是车里所有的人都饿了：意义仅限于车里的饥饿的人。）

In the second example, there are only two IPs, as opposed to three in the first. The subject noun phrase is sufficiently long to form a separate IP, but the restrictive relative clause within that noun phrase does not form a separate IP. The difference in meaning between (31) and (32) is that, in (32) it is not necessarily the case that all of the guys in the car were hungry: the meaning is restricted only to the hungry guys in the car.)

同位名词短语

Noun phrases in apposition

名词短语在共同指代时被称为同位词，也就是说，当它们被用来指代同一个人或实体时，例如：

Noun phrases are said to be in apposition when they are co-referential, that is, when they are being used to refer to the same person or entity, as in:

gueqrecard_002

(33) ˈBarak O↘ ba ma, | 一位民主党政治家， | 不聪明。（曲目 10.29）

其他括号

Other parentheticals

(34) ↘↗玛丽，| 你不会相信这个，| 但是简↗↘怀孕了！（曲目 10.30）

请注意，与前面和后面的 IP 相比，括号中的音高范围较低：如果您仔细聆听音轨 10.28、10.29和10.30，您应该能够听到这一点。

Note that parentheticals are uttered on a lower pitch range than the preceding and following IPs: if you listen carefully to Tracks 10.28, 10.29 and 10.30, you should be able to hear this.

10.3.1.2 协调的成分

10.3.1.2 Co-Ordinated Constituents

(35)	（一种）	ˈMary ˈ搬到了↘巴黎\| 但 ˈJohn 留在 ↘伦敦。（句子协调）
	（乙）	ˈJohn ˈ去了↘酒吧 \| 并点了一杯 ↘啤酒。（动词短语协调）
	（C）	她'非常↘高\| 并且非常↘漂亮。（形容词短语协调）
	(四)	他的'very ↘ well \| 并且非常 ↘很快。（副词短语协调）
	(五)	它要么在↘冰箱里\|要么或 ˈ在 ↘表上。（介词短语协调）
	（F）	他在↘山上买了房子\| 和↘山谷中的树林。（名词短语协调）（轨道 10.31）

gueqrecard_002

但是，当成分很短时，并不总是需要单独的 IP：

However, when the constituents are short, separate IPs are not always required:

(36)	（一种）	她又高又瘦。（形容词短语协调）
	（乙）	他停了下来，↘盯着看。（动词短语协调）
	（C）	他 ˈbought ˈmilk 和 ↘ cheese。（名词短语协调）（轨道 10.32）

当并列元素已经词汇化（形成词汇项目）时，这一点尤其明显，被用来指代单个实体，例如：

This is especially noticeable when co-ordinated elements have been lexicalized (formed into lexical items), taken to be used to refer to single entities, such as:

(37)	（一种）	ˈfish'n' ↘薯条
	（乙）	ˈ啤酒和 ↘吃喝玩乐
	（C）	ˈ草莓和 ↘奶油（曲目 10.33）

用于指代被视为单个实体或单位的其他协调项目是英国酒吧名称：

Other co-ordinated items that are used to refer to what is perceived as a single entity or unit are British pub names:

(38)	（一种）	The ˈDog and ↘鸭子
	（乙）	狐狸和 ↘猎犬

和夫妻的名字：

and names of couples:

gueqrecard_002

（C）

ˈBill 和 ↘ Mary ˈJane 和 ↘ Clive（轨道 10.34）

（这里的夫妻被认为是“一个项目”。）

10.3.1.3 清单上的项目

10.3.1.3 Items on Lists

通常，列表中的每个项目构成一个单独的音调组

Normally, each item on the list constitutes a separate tone group

(39) 他买了 ↗个鸡蛋，| ↗牛奶, | 到↗ ma脚趾| 和↘火腿。（曲目 10.35）

列表中的非最终项目通常采用升调：这表示列表尚未完成。

Non-final items on a list often take a rising tone: this signals that the list is not yet complete.

10.3.1.4 从句

10.3.1.4 Subordinate Clauses

当一个句子包含从句时，从句边界通常对应一个 IP 边界：

When a sentence contains a subordinate clause, the clause boundary often corresponds to an IP boundary:

(40)	（一种）	我要买 ˈfish'n' ↘薯条 \| 当我去 ↘商店时。
	（乙）	我告诉了↘公司的新招聘\| 他被 ↘解雇了。（曲目 10.36）

如果从句前面的材料比较短，则从句不需要单独构成一个IP：

If the material preceding the subordinate clause is relatively short, the subordinate clause need not form a separate IP:

gueqrecard_002

(41) 我认为她被 ↘解雇了。（曲目 10.37）

10.3.1.5 句子状语

10.3.1.5 Sentence Adverbials

副词短语必然具有副词功能。但是其他短语，尤其是介词短语，可以具有状语功能。

Adverb phrases necessarily have an adverbial function. But other phrases, notably prepositional phrases, can have an adverbial function.

区分动词短语状语和句子状语是很常见的，分别如（42a）和（b）所示：

It is common to distinguish verb phrase adverbials from sentence adverbials, as in (42a) and (b) respectively:

(42)	（一种）	约翰满怀希望地去面试了。
	（乙）	约翰满怀希望地去面试了。

在 (42a) 中，约翰满怀希望。在 (42b) 中，说话者充满希望。

In (42a), John was hopeful. In (42b), the speaker is hopeful.

请注意 (42b) 可以用句首的状语改写：

Notice that (42b) can be rephrased with the sentence adverbial in initial position:

（C）

希望约翰去面试了。

在任何一种情况下，句子状语都形成一个单独的 IP，并且它们具有降调：

In either case, sentence adverbials form a separate IP, and they have the fall-rise tone:

(43)	（一种）	↘↗满怀希望，\| ˈJohn ˈ去↘面试。
	（乙）	ˈJohn ˈ去↘面试， \| ↘↗满怀希望。（曲目 10.38）

10.3.1.6 伪裂缝

10.3.1.6 Pseudo-Clefts

这些采用句法形式他需要的是洗澡。这些通常会形成两个独立的 IP：

These take the syntactic form What he needs is a bath. It is common for these to form two separate IPs:

gueqrecard_002

(44) 他↘↗需要什么| 是↘浴。（曲目 10.39）

10.3.1.7是。. . 是那个建筑

10.3.1.7 The is . . . is that Construction

(45)	（一种）	↘↗是\| 是她 ↘怀孕了。
	（乙）	替代方案 ↘↗是\| 是我们必须更多地参与↘ Eu rope。
	（C）	我关心的↘↗是\| 是它太大↘了\| 太快了↘ 。
	(四)	↘↗ 的事实是\| 是↘运行的方式吗？太 ↘复杂了。
	(五)	“好”新闻↘↗是\| 是他们向小型 ↘企业提供贷款。（曲目 10.40）

gueqrecard_002

这种结构在英语口语中非常普遍，无论是非正式的还是正式的（这些例子大多来自 BBC 电视台对英国政客的正式采访）。虽然在第一个is之后可能没有 IP 边界，但这是常态。

This construction is very widespread in spoken English, both informal and formal (most of these examples are taken from formal interviews with British politicians on BBC TV). Although it is possible not to have an IP boundary after the first is, that is the norm.

10.3.2 通常不构成独立 IP的句法单元

10.3.2 Syntactic Units which Do Not Normally Form Separate IPs

10.3.2.1 报告条款

10.3.2.1 Reporting Clauses

这些在小说中比比皆是，但在日常谈话中也比比皆是

These abound in novels, but also in everyday speech

(46) 我 ↘累了，他说。（曲目 10.41）

在这里，主音节的降调一直拖到报告子句中（他说）

Here, the falling tone on the tonic syllable keeps trailing off into the reporting clause (he said)

但是，可以围绕报告条款形成单独的 IP。比较 (47a) 和 (b)：

It is possible, however, to form a separate IP around a reporting clause. Compare (47a) and (b):

(47)	（一种）	我想他是再红了。
	（乙）	他重新↘ ti红色，\| 我 ↘ th墨水。（曲目 10.42）

话语 (47b) 传达的确定性不如 (47a)。

Utterance (47b) conveys less certainty than (47a).

10.3.2.2 主语名词短语

10.3.2.2 Subject Noun Phrases

gueqrecard_002

(48) 老大爷踢了↘狗。（曲目 10.43）

但是，正如我们之前看到的，主题名词短语越长，独立 IP 的可能性就越大。

But, as we saw in earlier, the longer a subject noun phrase, the more likely it is that a separate IP will be possible.

10.3.2.3 主语名词短语中的限制性关系从句

10.3.2.3 Restrictive Relative Clauses within Subject Noun Phrases

正如我们所见，这些通常不会形成一个单独的 IP，如示例 (32) 所示，此处重复（长主语 NP 形成一个单独的 IP，但不是关系从句）：

As we have seen, these do not normally form a separate IP, as in example (32), repeated here (the long subject NP forms a separate IP, but not the relative clause):

(32) ↘饿了|车里的伙计们 ˈ吃了一些↘三明治。

10.3.3 句法单元可能会或可能不会形成单独的 IP，这取决于所传达的含义

10.3.3 Syntactic Units which May, or May Not, Form Separate IPs, Depending on the Sense Conveyed

10.3.3.1 标记问题

10.3.3.1 Tag Questions

反极性标签问题

Reverse polarity tag questions

通过反极性，我们的意思是第一部分是正面的，标签问题是负面的，反之亦然，如：

By reverse polarity we mean that the first part is in the positive, and the tag question in the negative, or vice versa, as in:

gueqrecard_002

(49)	（一种）	你会这样做的，不是吗？
	（乙）	你不会这样做的，是吗？（曲目 10.44）

反极性标签问题的语调是这样的：如果我们在标签问题上形成一个单独的IP，并在其上放置降调，则标签邀请同意：

The intonation of reverse polarity tag questions works as follows: if we form a separate IP on the tag question, and place a falling tone on it, the tag invites agreement:

(50)	（一种）	你要 ↘做这个，\| ↘不是吗？
	（乙）	你不会 ↘这样做，\| ↘你是？（曲目 10.44）

如果tag上有升调，就不用单独IP了：

If there is a rising tone on the tag, it need not form a separate IP:

(51)	（一种）	她“要来参加 ↗派对了，不是吗？
	（乙）	她不会“结婚”，是吗？（曲目 10.45）

在这里，标签问题构成了 IP 尾部的一部分。但是标签问题可以形成一个单独的IP：

Here, the tag question forms part of the tail of the IP. But the tag question can form a separate IP:

gueqrecard_002

(52)	（一种）	她''来参加↗派对，\| ↗不是她吗？
	（乙）	她不是'getting ↗ married , \| ↗是她吗？（曲目 10.46）

传达的意思的差异是微妙的：在 (51a) 中，说话者不完全确定她是否要来参加聚会。在 (51b) 中，说话者可能表达了惊讶，甚至是惊讶，而在 (52a) 和 (52b) 中，说话者不太确定，并且比 (51a) 和 (51b) 提出了更多的疑问.

The differences in conveyed meaning are subtle: in (51a), the speaker is not entirely certain whether she’s coming to the party. In (51b), the speaker may be expressing surprise, or even astonishment, whereas in (52a) and (52b), the speaker is a lot less sure, and is posing more of a query than in (51a) and (51b).

10.4 主音放置、IP 边界和语法

10.4 Tonic Placement, IP Boundaries and Syntax

10.4.1 短语动词

10.4.1 Phrasal Verbs

英语中的短语动词有两部分：第一部分，看起来像普通动词；第二部分，看起来像介词，通常称为助词。它们可以是及物的（这意味着它们后面跟着一个直接宾语，如在He chated up the waitress中，女服务员是直接宾语）或不及物的（这意味着不需要直接宾语，如在He backed down中） . 学习者作为外语的英语学习者最好学习这些动词的语调，因为它们的数量太多，而且它们在英语口语中出现的频率很高。

Phrasal verbs in English have two parts: the first part, which looks like a normal verb, and the second part, which looks like a preposition, and is often called a particle. They can be transitive (which means that they are followed by a direct object, as in He chatted up the waitress, where the waitress is the direct object) or intransitive (which means that no direct object is required, as in He backed down). Learners of English as a foreign language are well-advised to learn the intonation of such verbs, since there are so many of them, and they occur with high frequency in spoken English.

10.4.1.1 及物短语动词

10.4.1.1 Transitive Phrasal Verbs

如果直接宾语名词短语是短语，则主音落在名词短语中的中心名词上（中心名词是名词短语中语义最突出的名词）：

If the direct object noun phrase is phrasal, the tonic falls on the head noun in the noun phrase (the head noun is the noun in a noun phrase which is semantically the most prominent):

(53)	（一种）	他在 ↘ wai tress 上聊天。
	（乙）	他在 ↘ wai tress 上聊天。

如果直接宾语名词短语是代词，则主音落在助词上（通常会移动以使其跟随直接宾语）：

If the direct object noun phrase is a pronoun, the tonic falls on the particle (which is normally shifted so that it follows the direct object):

gueqrecard_002

10.4.1.2 不及物短语动词

10.4.1.2 Intransitive Phrasal Verbs

这些采取粒子上的补品：

These take the tonic on the particle:

(54) He ˈbacked ↘ down . （曲目 10.48）

然而，许多带有不及物短语动词的短句是“事件句”（见上文 10.2.4），在这种情况下，主音被收回：

However, many short sentences with intransitive phrasal verbs are ‘event sentences’ (see 10.2.4 above), in which case the tonic is retracted:

(55)	（一种）	↘飞机爆炸了。
	（乙）	↘汽车抛锚了。（曲目 10.49）

10.4.2 程度状语

10.4.2 Degree Adverbials

程度状语最主要的例子是very这个词。它的作用是修饰形容词短语中的形容词和副词短语中的副词，例如：

The most central example of a degree adverbial is the word very. It functions to modify adjectives in adjective phrases, and adverbs in adverb phrases, as in:

(56)	（一种）	他非常↘高。（形容词短语）
	（乙）	他说话很慢↘ 。（副词短语）（曲目 10.50）

其他程度状语包括so , incredibly，温和的脏话bloody和更强的 'f-word'，如：

Other degree adverbials include so, incredibly, the mild swearword bloody and the stronger ‘f-word’, as in:

gueqrecard_002

(57)	（一种）	他真是 ↘笨蛋。
	（乙）	他是一个令人难以置信的 ↘傲慢。
	（C）	他是“该死的 ↘好人” 。
	(四)	他真他妈的好↘ 。（曲目 10.50）

但是补品可以放在程度状语上，以强调，例如：

But the tonic can be placed on the degree adverbial, for emphasis, as in:

gueqrecard_002

(58)	（一种）	他↗↘太傻了
	（乙）	他令人信服↗↘狂妄！
	（C）	他是 ↗↘ bloo dy
	(四)	他 ↗↘f*cking 得好！（曲目 10.51）

so在口语中用作动词短语状语，带有感叹语气的补品，如：

There is a use of so in colloquial English which acts as a verb phrase adverbial, and takes the tonic, with an exclamatory tone, as in:

(59)

10.5 声调和句法

10.5 Tones and Syntax

10.5.1 WH 问题

10.5.1 WH Questions

这些通常有降调：

These normally have a falling tone:

gueqrecard_002

(60) Where are you ↘ going ing ? （曲目 10.53）

但不是在回声使用时：

But not when used echoically:

(61)	A：	我要搬到 ↘伦敦。
	乙：	↗你要搬到哪里？（曲目 10.54）

这里的说话者 B 要么没有正确听到 A 所说的话，要么表示怀疑。B 的 WH 问题被认为是重复的，因为它重复了 A 刚才所说的部分或全部内容。请注意，在 IP 的尾部，声调继续向上拖尾，就像当尾部之前的主音音节具有降调时它向下拖尾一样。

Speaker B here has either not properly heard what A said, or is expressing incredulity. B’s WH question is said to be echoic in that it echoes part or all of what A has just said. Notice that the tone keeps on trailing upwards in the tail of the IP, just as it trails downwards when a tonic syllable before the tail has a falling tone.

10.5.2 陈述式疑问句

10.5.2 Declaratives as Questions

我们可以使用句法来形成是/否问题，其中主句中的第一个助动词倒置在主语名词短语周围，如：

We can use syntax to form yes/no question, in which the first auxiliary verb in the main clause is inverted around the subject noun phrase, as in:

(62) (a) 他们找到你的 ↗手机了吗？

相应的声明语句将具有降调：

The corresponding declarative statement would have a falling tone:

(b) 他们找到了您的 ↘手机。

但是我们可以保留声明句法结构，并通过在 LLI 上放置一个升调而不是降调来提出问题：

But we can retain the declarative syntactic structure and still ask a question by placing a rising, rather than a falling, tone on the LLI:

gueqrecard_002

陈述式结构可以用下降 - 上升来表达：

Declarative structures can be uttered with a fall-rise:

(63)

在使用 fall-rise 时，B 表达了对 mobile 已经找到的惊讶，或者是mobile，而不是其他一些丢失的物体，已经找到了。

In using a fall-rise, B is expressing surprise that the mobile has been found, or that it is the mobile, as opposed to some other lost object, that has been found.

10.6 主音位置和语篇语境

10.6 Tonic Placement and Discourse Context

10.6.1 呼格

10.6.1 Vocatives

呼唤表达用于称呼一个人的对话者，如以下摘录自一位演讲者打电话给一位名叫尼克的朋友留下的电话留言：

Vocative expressions are used for addressing one’s interlocutor, as in the following extracts from telephone messages left by a speaker calling a friend called Nick:

(64)	（一种）	尼克，是我。
	（乙）	是我，尼克。

这里说话者假设收件人（尼克）可以识别呼叫者的声音，或者尼克正在等待那个人打来的电话。

Here the speaker assumes that the addressee (Nick) can identify the voice of the person calling, or that Nick is expecting a call from that person.

初始呼格形成一个单独的 IP：

Initial vocatives form a separate IP:

最终呼格不：

Final vocatives do not:

gueqrecard_002

(d) ↘我，尼克。（曲目 10.56）

如果 IP 围绕最后的Nick形成，如 (65) 所示：

If an IP is formed around the final Nick, as in (65):

(65) 是↘我，| ↘尼克。（曲目 10.57）

那么尼克这个词不被解释为呼格；相反，它被解释为调用者的名字。

then the word Nick is not interpreted as a vocative; rather, it is interpreted as the name of the caller.

10.6.2 IP边界传达的其他含义差异

10.6.2 Other Meaning Differences Conveyed by IP Boundaries

正如我们所见，IP 边界的位置和/或我们选择的语气类型可以传达不同的含义。考虑：

As we’ve seen, the placement of IP boundaries, and/or the kind of tone we select, can convey differences in meaning. Consider:

(66)

（一种）

他娶她不是因为她是法国人。

(= He ˈdidn't ˈmarry her because she was ↘ French )

这意味着他确实娶了她，但不是因为她是法国人。

This means that he did marry her, but not because she was French.

gueqrecard_002

（乙）

他没有娶她，因为她是法国人。( Track 10.58 )

(= He ˈdidn't ↘ marry her | because she was ↘ French )

这意味着他没有娶她，原因是她是法国人。

This means that he didn’t marry her, the reason being that she was French.

10.7 总结

10.7 Summing Up

我们已经看到英语语调的三个主要结构方面：将话语分成作为信息块的语调短语，在每个块中的一个重读音节上放置一个补品，以及我们使用的语调类型在那个补品音节中。我们已经看到，英语中的语调与句法结构、词汇与功能的区别、所讨论的句法单元所表达的含义以及与话语语境和现象（如传达的含义和说话者的态度）相关的话语方面有关对她或他所说的话。也许英语语调最引人注目的方面是它的动态程度，因为说英语的人经常出于各种目的将主音从默认的 LLI 位置移开。

We have seen that there are three main structural aspects of English intonation: the dividing up of utterances into intonational phrases which are chunks of information, the placing of a tonic on one of the stressed syllables in each chunk, and the kind of tone we use in that tonic syllable. Intonation in English, we have seen, is connected to syntactic structure, the lexical vs functional distinction, the meaning expressed by the syntactic units in question, and aspects of discourse linked to the context of utterance and phenomena such as conveyed meaning and the speaker’s attitude towards what she or he is saying. Perhaps the most striking aspect of English intonation is the extent to which it is dynamic, in the sense that speakers of English frequently move the tonic away from the default LLI position, for a wide variety of purposes. For non-native speakers of English, some degree of mastery of this will result in a much more native-like speech style.

练习

Exercises

1 在以下每个话语中，找出最后一个词项。

(a) 约翰去了酒吧。

(b) 玛丽把她的手指放在上面。

(d) 他说话很慢。

(e) 我想要那个粉红色的。

2 以下哪些问题可以用升调？

(a) 布什疯了吗？

(b) 你想要什么？

（三）你吃了吗？

(d) 这是如何运作的？

(e) 是不是该吃午饭了？

gueqrecard_002

3 收听曲目 10.59。该声音文件中以下话语中的补品在哪里？解释为什么。

(a) 她和服务员聊天。

(b) 她和他搭讪。

(d) 我们分手了。

(e) 我让他失望了。

4 收听曲目 10.60。识别该声音文件中以下话语中的 IP 边界，并说出主音落在哪里。解释为什么。

(a) 玛丽，你被解雇了。

(b) 你被解雇了，玛丽。

(d) 美国总统说，这是一个邪恶的帝国。

5 收听曲目 10.61。识别该文件中以下话语中的 IP 边界，并说出补品落在哪里。解释为什么。

(a) 我的好朋友玛丽怀孕了。

(b) 车上那些饿了的人吃了一些三明治。

(d) Bill，你不会相信的，你已经通过了考试。

(e) 他的新书《交朋友》肯定会成为畅销书。

6 以下话语中的默认主音位置是什么？

(a) 他星期四去了伦敦。

(b) 我最近没见过她。

(d) 她把卧室弄得一团糟。

(e) 他说话很快。

gueqrecard_002

7 下列话语中的语调组有哪些可能性？讨论标签问题中不同语气的可能性。听曲目 10.62 并描述您听到的语调结构，包括 IP 边界和音调。

(a) 你没有怀孕，是吗？

(b) 你喜欢烤宽面条，是吗？

(d) 我们可以解决这个问题，不是吗？

(e) 我们永远不会解决这个问题，对吗？

gueqrecard_002

8 以下话语中的声调边界在哪里，补音在哪里？如果有其他语调结构，请说出您的想法。聆听Track 10.63并描述您听到的语调结构，包括 IP 边界和音调。

(a) “你不能去”，比尔史密斯说，他是我的好朋友。

(b) “奥巴马无法在得克萨斯州获胜”，希拉里·克林顿 (Hillary Clinton) 声称，她的丈夫、前总统比尔来自美国南部。

(d) “Dickens I can't stand”，这位年轻的应聘者在维多利亚时代文学的大学讲师中坦白道。

(e) “虔诚的基督徒乔治布什对阿布格莱布监狱的囚犯待遇有何看法？”委员会主席问道。

11 字音学：拼写-发音关系

Graphophonemics: Spelling–Pronunciation Relations

11.1 简介

11.1 Introduction

英语拼写和发音之间的关系很复杂：它看起来完全是任意的。复杂性有历史渊源：它们源于诺曼征服后抄写员引入的变化，采用许多外来拼写的外来词，最重要的是，随着英语从古英语演变为中世纪英语，语音系统发生了变化, 分为早期现代英语和现代英语。尽管很复杂，但还是有一些基本规律值得学习，尤其是对于非英语母语者。

The relationship between spelling and pronunciation in English is complex: it can seem completely arbitrary. The complexities have historical sources: they result from changes introduced by scribes after the Norman Conquest, the adopting of many loanwords with foreign spellings, and, above all, changes in the phonological system of English as the language evolved from Old English, through Middle English, into Early Modern English and present-day English. Despite the complexities, there are some basic regularities which are worth learning, especially for non-native speakers of English.

我们首先区分字母和字形。罗马字母表中有 26 个字母，但有超过 26 个视觉符号（字素）用于表示英语音素和音位变体，因为可以使用字母组合来表示给定的音素或音位变体。例如，<ph> 对应于 /f/ 音素，如单词photograph，<th> 对应于 /θ/ 和 /ð/ 音素，如单词think和this，以及 <oa >，对应于 RP 和 GA 音素 /oʊ/ 以及 SSE（标准苏格兰英语）音素 /o/，如单词boat中。我们将此类字素称为有向字母，因为它们包含两个字母。我们将在诸如schmaltzy（“过度多愁善感”）之类的词中将具有三个字母（例如 <sch>）的字素称为三字母组。字母和字形之间的区别可以从英语和法语的不同书写习惯中看出。例如，在用英文写我名字的首字母时，惯例是使用第一个字母：'P。卡尔'。法语惯例是选择第一个字素：'Ph. 卡尔'。字音组学是研究字素和音素（以及它们的一些音位变体）之间关系的名称。我们将从检查英语中的元音字素开始，然后继续研究辅音字素。

We begin by distinguishing between letters and graphemes. There are twenty-six letters in the Roman alphabet, but there are more than twenty-six visual symbols (graphemes) for representing English phonemes and allophones, since combinations of letters can be used to represent a given phoneme or allophone. Examples are , which corresponds to the /f/ phoneme, as in the word photograph, , which corresponds to both the /θ/ and /ð/ phonemes, as in the words think and this, and , which corresponds to the RP and GA phonemes /oʊ/ and the SSE (Standard Scottish English) phoneme /o/, as in the word boat. We will refer to such graphemes as digraphs, since they contain two letters. We will refer to graphemes with three letters, such as , in words such as schmaltzy (‘overly sentimental’), as trigraphs. The distinction between letters and graphemes can be seen in the different writing conventions of English and French. For instance, in writing the initial for my first name in English, the convention is to use the first letter: ‘P. Carr’. The French convention is to select the first grapheme: ‘Ph. Carr’. The term graphophonemics is the name given to the study of the relationship between graphemes and phonemes (and some of their allophones). We will begin by examining vowel graphemes in English, and then proceed to consonant graphemes.

11.2 元音字素及其音位值

11.2 Vowel Graphemes and Their Phonemic Values

11.2.1 元音专论

11.2.1 Vowel Monographs

让我们从五个元音字母 <a>、<e>、、<o>、 开始。我们将区分这些字素的两个不同音素值：它们的检查值和它们的自由值。从历史上看， checked和free这两个词源自有关音节结构的事实。以bite这个词为例：在中古英语中，这个词是双音节的：/bi:tə/，其中第一个音节是 /bi:/，第二个音节是 /tə/。它们都是开音节：它们不包含尾辅音。从历史上看，我们所说的自由价值源自开放的音节。字位, 在中古英语中，发音为 /bit/，一个闭音节的单音节词，包含尾辅音 /t/。我们所说的检查值源自闭音节：闭音节是检查音节。请注意，在现代英语中，单词bite失去了最后的 schwa，现在是单音节词：/baɪt/。但它保留了一个自由值：/aɪ/，长元音 /i:/ 的历史后裔。单词bit保留了一个检查值：/ɪ/，短元音 /i/ 的历史后裔。

Let us begin with the five vowel graphemes , , , , . We will distinguish between two different phonemic values for these graphemes: their checked values and their free values. The terms checked and free derive, historically, from facts about syllable structure. Take the word bite: in Middle English, this word was bisyllabic: /bi:tə/, in which the first syllable is /bi:/, and the second syllable is /tə/. These are both open syllables: they contain no coda consonants. What we are referring to as free values derive, historically, from open syllables. The word bit, in Middle English, was pronounced /bit/, a monosyllabic word with a closed syllable, containing the coda consonant /t/. What we are referring to as checked values derive from closed syllables: a closed syllable is a checked syllable. Notice that, in contemporary English, the word bite has lost the final schwa, and is now monosyllabic: /baɪt/. But it retains a free value: /aɪ/, the historical descendant of the long /i:/ vowel. The word bit retains a checked value: /ɪ/, the historical descendant of the short /i/ vowel.

在我们这里要考虑的单音节词种类中，其中四个字素有一个checked和一个free值，而有两个checked和free值，如下（来自RP的例子）：

In the kinds of monosyllabic words we will consider here, four of these graphemes have a checked and a free value, and has two checked and free values, as follows (examples from RP):

(1)

虽然最后的 <e> 保留在第一列示例中的单词拼写中，但它曾经对应的 schwa 随着时间的推移被省略了，正如我们已经注意到的那样。我们将这个字素称为静音 e。在重音单音节词中，如果有一个不发音的e，则前元音字素对应自由值。请注意，这些字素的名称具有自由值：我们称它们为 /eɪ/、/i:/、/aɪ/、/oʊ/、/ju:/。

While the final remains in the spelling of the words in the first column of examples, the schwa which it once corresponded to was elided over time, as we have noted. We will refer to this grapheme as mute e. In stressed monosyllabic words, if there is a mute e, the preceding vowel grapheme corresponds to the free value. Note that the names of these graphemes have the free value: we call them /eɪ/, /i:/, /aɪ/, /oʊ/, /ju:/.

在没有哑音 e 的重读单音节词中，前面的元音字素对应于选中的值。当书面单词末尾有多个辅音字素时，情况就是如此，例如：

In a stressed monosyllabic word without a mute e, the preceding vowel grapheme corresponds to the checked value. This is true when there is more than one consonant grapheme at the end of the written word, as in:

(2)	恰当，行动
	铃声，文字
	票据，宽度
	搭讪，结
	屁股，簇绒

当某些词缀被添加到双音节词时，这些自由/检查值代表发音的交替，如以下来自 RP 的示例：

These free/checked values figure in alternations in pronunciation when certain affixes are added to bisyllabic words, as in the following examples, from RP:

(3)

在重读的单音节词中，这些元音字素在 RP 等非变音重音中有额外的值，其中 [ɹ] 在尾音位置的历史损失导致前面的元音发生变化：

In stressed monosyllabic words, there are additional values for these vowel graphemes in non-rhotic accents such as RP, where the historical loss of [ɹ] in coda position has resulted in changes to the preceding vowel:

(4) RP 中的“Pre-r”检查和自由值

在当代RP中，sure这样的词的/(j)ɔ:/读音似乎正在慢慢取代之前的/(j)ʊə/读音，所以shore和sure是同音字。许多像mare这样的词现在发音为 /ε:/，而不是 /εə/。/aɪə/ 的单音版本在 RP 中有一个长 [a:]：RP 说话者之间存在差异。

In contemporary RP, the /(j)ɔ:/ pronunciation of words like sure seems to be slowly replacing the previous /(j)ʊə/ pronunciation, so that shore and sure are homophones. Many words like mare are now pronounced /ε:/, rather than /εə/. Monophthongal versions of /aɪə/ have a long [a:] in RP: there is variability here among RP speakers.

正如我们所见，GA 是 rhotic，因此“pre-r”位置的自由值和选中值与 RP 值有些不同（为清楚起见，我们包括 /ɹ/）：

As we have seen, GA is rhotic, so the free and checked values in ‘pre-r’ position differ somewhat from the RP values (we include the /ɹ/ for clarity):

(5) GA 中的'Pre-r' checked and free values

在具有多个音节且最后一个音节被重读的单词中可以找到相同的检查值和自由值模式（来自 RP 的示例）：

The same pattern of checked and free values can be found in words with more than one syllable with the final syllastressed (examples from RP):

(6)

对于具有多个音节的单词中的“pre-r”元音字素也是如此，重读在最后一个音节上：

The same is true for ‘pre-r’ vowel graphemes in words with more than one syllable, stressed on the final syllable :

(7)

我们已经看到几个带有重读末音节和单辅音字素且没有静音 e 的单词示例，例如mad、pet、hid、not、cut、forget、forbid、uncut '：重读音节中的元音字素对应于字素的检查值。如果我们为这些词添加屈折后缀，从而添加一个音节，我们必须加倍辅音字素来传达检查值：madden、petting、hidden、knotting、cutting、forgetged、forbidden。从某种意义上说，英语单词的拼写实际上有助于指导发音：如果没有双辅音字素，重读元音字素将具有自由值，如下对：

We have seen several examples of words with a stressed final syllable and a single consonant grapheme with no mute e, such as mad, pet, hid, not, cut, forgot, forbid, uncut’: the vowel grapheme in the stressed syllable corresponds to the checked value of the grapheme. If we add inflectional suffixes to such words, and thus add a syllable, we must double the consonant grapheme to convey the checked value: madden, petting, hidden, knotting, cutting, forgotten, forbidden. This is one sense in which the spelling of English words can actually be helpful as a guide to pronunciation: without the doubled consonant grapheme, the stressed vowel grapheme would have the free value, as in the following pairs:

(8)

上面的词在形态上很复杂：它们包含不止一个语素。但即使是形态简单且词前有重读音节的单词，我们也发现了相同的模式。例如，单词hammer不包含后缀，但双辅音字素编码 <a> 的检查值，即 /æ/。laser一词在形态上也很简单，但单个辅音字素编码 <a> 的自由（长/时）值，即 /eɹ/（在 RP 和 GA 中）。不幸的是，有不规则的拼写。拿字面板来说，它具有检查过的 /æ/ 值：我们希望它与 /eɹ/ 一起发音，或者写成 pannel ，就像flannel和channel一样。这些不规则的拼写对外国学习者没有帮助，他们经常在学习发音之前先学习单词的拼写。但是我们刚才描述的规律性仍然涵盖了大量的英语单词，因此值得学习。

The words above are morphologically complex: they contain more than one morpheme. But even in words which are morphologically simple and have a prefinal stressed syllable, we find the same pattern. The word hammer, for instance, does not contain a suffix, but the double consonant grapheme encodes the checked value for , namely /æ/. The word laser is also morphologically simple, but the single consonant grapheme encodes the free (long/tense) value for , namely /eɹ/ (in RP and GA). Unfortunately, there are irregular spellings. Take the word panel, which has the checked /æ/ value: we would expect it to be pronounced with /eɹ/, or to be written pannel, just like flannel and channel. These irregular spellings do not help the foreign learner, who often learns the spelling of a word before learning the pronunciation. But the regularities we have just described nonetheless cover a huge number of English words, and are therefore worth learning about.

更进一步的规律涉及倒数第二个音节中的重读元音（或比那个更早的单词），当它们后面跟着一个辅音字素时：这些通常具有检查值，如以下单词（RP 发音）：

A further regularity concerns stressed vowels in antepenultimate syllables (or earlier in the word than that) when they are followed by a single consonant grapheme: these typically have the checked value, as in the following words (RP pronunciations):

(9)

字素的行为不同：倒数第二个位置（或更早）的重读元音，当其后跟单个辅音字素时， 具有自由值，如：

The grapheme behaves differently: a stressed vowel in penultimate position (or earlier), when followed by a single consonant grapheme, has the free value for , as in:

(10)	节俭	/ˈfɹu:gəl/
	轻率的	/ɪmˈpɹu:dənt/
	积累	/əˈkju:mjəleɪt/

除了<a>、<e>、、<o>、专着外，<y>字素还可以表示元音，如下：

In addition to the , , , , monographs, the grapheme can represent vowels, as follows:

(11)

字素	音素/音位变体	例子
<y>	/ɪ/, /aɪ/, [i]	神话，押韵，快乐

在重读音节中，<y> 与 具有相同的值。检查值 /ɪ/ 出现在我们期望的位置：像myth这样的词，由一个重读音节组成，没有静音 e，而像rhyme这样的词有静音 e，因此有自由值 /aɪ /. 我们已经看到，在'pre-r'位置，当后面跟一个静音 e 时，RP 有 /aɪə/，就像在tire中一样，可以发长 [a:]。这个位置的<y>也是一样：tire对应/taɪə/，也可以读成[ta:]。

In stressed syllables, has the same values as . The checked value /ɪ/ occurs where we would expect it to: words like myth, consisting of a single stressed syllable, do not have a mute e, whereas words such as rhyme have a mute e, and thus have the free value /aɪ/. We have seen that, in ‘pre-r’ position, when followed by a mute e, RP has /aɪə/, as in tire, which can be uttered as long [a:]. The same is true for in this position: tyre corresponds to /taɪə/, which can also be pronounced [ta:].

对于词尾非重读 <y>，我们将此处的短元音 [i] 视为 RP 和 GA 中 /i:/ 音素的位置变体（异音），它出现在诸如happy之类的词中，并出现在许多以 <ly> 结尾的副词，例如lovely。重读词尾 <y> 对应于 /aɪ/，如dry和fly。

For word-final unstressed , we have treated the short [i] vowel here as a positional variant (allophone) of the /i:/ phoneme in RP and GA, where it occurs in words such as happy, and in the many adverbs ending in , such as lovely. Stressed word-final corresponds to /aɪ/, as in dry and fly.

11.2.2 元音二合字母

11.2.2 Vowel Digraphs

元音二合字母给外国学习者带来了问题，因为出于与英语历史相关的原因，它们通常对应于多个音位值。对于 RP 尤其如此，因为它是非流变性的，这使事情变得复杂。让我们列出一些最常见的元音二合字母并讨论它们的音位值。

The vowel digraphs present problems for foreign learners, since they often correspond to more than one phonemic value, for reasons connected to the history of English. This is especially true for RP, since it is non-rhotic, which complicates matters. Let us list some of the most frequent vowel digraphs and discuss their phonemic values.

11.2.2.1 <人工智能>

11.2.2.1

在 SSE 中，<ai> 对应音素 /e/，如pain (/pen/) 和fair (/feɹ/)。但是，在 RP 中，由于它是非 Rhotic 的，因此有一个单独的“pre-r”值：pain是 /peɪn/，而fair是 /fεə/ 或 /fε:/。在这方面，对于许多英语学习者来说，SSE 不如 RP 复杂。在 GA 中，有一个复杂的问题：pain是 /peɪn/，就像在 RP 中一样，但是像fair这样的 'pre-r' 词，虽然有 /eɪ/ 音素，却经历了 'Marry Merry Mary' 中和，其中 / æ/ vs /ε/ vs /eɪ/ 在‘pre-r’位置被中和为 [ε]：Marry Merry Mary发音为 [ˈmεɹi] [ˈmεɹi] [ˈmεɹi]，公平发音为 [fεɹ]。在听美国人说话时，这可能会使说各种英式英语的人感到困惑。例如，当我的儿子在美国读小学一个学期时，告诉我一个新的美国同学的名字，我以为这个男孩叫Ferl，因为我儿子的发音和普通话的发音一样: [fεɹɫ]。事实上，男孩的名字是Farrel '：schwa 在 GA 中 /ɹ/ 之后经常被省略，并且“Marry Merry Mary”规则适用于重读元音。正如剧作家萧伯纳曾经说过的，英美是被同一种语言分割开来的两个国家！

In SSE, corresponds to the phoneme /e/, as in pain (/pen/) and fair (/feɹ/). But, in RP, since it is non-rhotic, there is a separate ‘pre-r’ value: pain is /peɪn/, while fair is /fεə/ or /fε:/. In this respect, SSE is less complex than RP for many learners of English. In GA, there is a complication: pain is /peɪn/, as in RP, but ‘pre-r’ words such as fair, while having the /eɪ/ phoneme, undergo the ‘Marry Merry Mary’ neutralization, in which the /æ/ vs /ε/ vs /eɪ/ oppositions are neutralized to [ε] in ‘pre-r’ position: Marry Merry Mary is pronounced as [ˈmεɹi] [ˈmεɹi] [ˈmεɹi], and fair is pronounced [fεɹ]. This can cause confusion for speakers of varieties of British English when listening to Americans. For instance, when my son, while attending primary school in the United States for a semester, told me the name of a new American schoolfriend, I thought that the boy was called Ferl, since my son pronounced the name the way a GA speaker would: [fεɹɫ]. In fact, the boy’s name was Farrel’: schwa is often elided after /ɹ/ in GA, and the ‘Marry Merry Mary’ rule applies to the stressed vowel. As the playwright George Bernard Shaw once said, Britain and America are two countries divided by a common language!

11.2.2.2 <au>

11.2.2.2

在 RP 中，这可以对应于 /ɔ:/，如在fraud中，或对应于 LOT 元音 /ɒ/，如在Austria中。许多说 GA 的人缺少 /ɒ/ 音素，而用 /ɑ/ 代替，因此 Austria 在其重读音节中具有 /ɑ/ 的值（尽管这里有可变性）。

In RP, this can correspond either to /ɔ:/, as in fraud, or to the LOT vowel /ɒ/, as in Austria. Many speakers of GA lack the /ɒ/ phoneme, and have /ɑ/ instead, so that Austria has the value /ɑ/ in its stressed syllable (though there is variability here).

11.2.2.3 <ee>

11.2.2.3

在 SSE 中，这对应于 /i/ 音素，如see (/si/) 和peer (/piɹ/)。在 RP 中，像see这样的词有 /i:/ 音素，而在“pre-r”位置，有一个居中的双元音，如peer (/piə/)。在GA中，<ee>对应于/i:/，但在'pre-r'位置往往会中和/i:/和/ɪ/之间的对立关系，使stir it and steer可以谐音。当这种对立被中和时，产生的元音听起来像 [ɪ] 或 [i]：[stɪɹɪt] 或 [stiɹɹɪt]。

In SSE, this corresponds to the /i/ phoneme, as in see (/si/) and peer (/piɹ/). In RP, words like see have the /i:/ phoneme, while, in ‘pre-r’ position, there is a centring diphthong, as in peer (/piə/). In GA, corresponds to /i:/, but there is often neutralization of the opposition between /i:/ and /ɪ/ in ‘pre-r’ position, so that stir it and steer it can be homophonous. When this opposition is neutralized, the resulting vowel can sound like [ɪ] or [i]: [stɪɹɪt] or [stiɹɹɪt].

11.2.2.4 <oo>

11.2.2.4

由于历史原因，这个二合字母在 RP 和 GA 中具有三个主要的音位值：/u:/，如soon中的，/ʊ/，如good中的，和 /ʌ/，如blood 中的。在 RP 的“pre-r”位置，中心双元音 /ʊə/ 出现，如在poor中，但这越来越多地让位给长单元音 /ɔ:/，因此sure和shore是同音字。在 GA 中，FOOT 元音可以不圆唇，通常用来表示产生的高后不圆唇元音的符号是 [ɯ]，如sure：[ʃɯɹ]。

For historical reasons, this digraph has three main phonemic values in RP and GA: /u:/, as in soon, /ʊ/, as in good, and /ʌ/, as in blood. In ‘pre-r’ position in RP, the centring diphthong /ʊə/ occurs, as in poor, but this is increasingly giving way to the long monophthong /ɔ:/, so that sure and shore are homophones. In GA, the FOOT vowel can be unrounded, and the symbol often used to represent that resulting high, back unrounded vowel is [ɯ], as in sure: [ʃɯɹ].

11.2.2.5 <你>

11.2.2.5

这个二合字母可以对应于 /aʊ/，如how或 /əʊ/ (GA /oʊ/)，如know。这给英语学习者带来了问题，因为没有规则来控制一个或另一个的出现。更糟糕的是，还有bow等形式，其中to bow的动词有 /aʊ/，而名词bow有 /əʊ/。

This digraph can correspond to either /aʊ/, as in how, or to /əʊ/ (GA /oʊ/), as in know. This poses problems for learners of English, since there is no rule governing the occurrence of one or the other. Worse still, there are forms such as bow, in which the verb to bow has /aʊ/, while the noun bow has /əʊ/.

在 <ought> 序列中，RP 有 /ɔ:/，如thought，而许多 GA 说话者在这些词中有 /ɒ/。

In sequences, RP has /ɔ:/, as in thought, while many GA speakers have /ɒ/ in such words.

词尾 <ough> 很复杂：它可以对应于 /u:/，如through，/əʊ/ (GA /oʊ/)，如though，/ʌf/，如tough，/ɒf/，如trough和 /aʊ/，如在犁中。

Word-final is complex: it can correspond to /u:/, as in through, /əʊ/ (GA /oʊ/), as in though, /ʌf/, as in tough, /ɒf/, as in trough, and /aʊ/, as in plough.

最后，<ou> 在带有 <ouble> 和 <ouple> 的单词中对应于 /ʌ/，如double和couple。

Finally, in words with and corresponds to /ʌ/, as in double and couple.

11.2.2.6 <ea>

11.2.2.6

对于英语作为外语的学生来说，这是最难的元音二合字母之一。它可以对应于 RP（和 GA）中的 /i:/，如单词sea或 /ε/，如head。RP 中有几个“pre-r”值：/ɪə/，如fear，/εə/（或者，在当代 RP 中更常见，/ε:/），如bear，/ɜ:/，如dearth , 和 /ɒ:/，如在hearth中。

This is one of the most difficult of the vowel digraphs for students learning English as a foreign language. It can correspond to /i:/ in RP (and GA), as in the word sea, or /ε/, as in head. There are several ‘pre-r’ values in RP: /ɪə/, as in fear, /εə/ (or, more often in contemporary RP, /ε:/), as in bear, /ɜ:/, as in dearth, and /ɒ:/, as in hearth.

在 GA 中，'pre-r' 值包括 /ɜ/，如dearth (/ dɜɹθ /) 和 /ɑ/，如hearth。由于“Marry Merry Mary”中和规则，GA 也有拼音 [ε]，如bear中。

In GA, the ‘pre-r’ values include /ɜ/, as in dearth (/dɜɹθ/), and /ɑ/, as in hearth. GA also has phonetic [ε], as in bear, because of the ‘Marry Merry Mary’ neutralization rule.

11.3 辅音字素及其音位值

11.3 Consonant Graphemes and Their Phonemic Values

11.3.1 辅音专着

11.3.1 Consonant Monographs

英语中有辅音专着，它们具有单一的音位值。这些都是：

There are consonant monographs in English which have a single phonemic value. These are:

(12)

字素	音素	例子
<p>	/p/	坑
<k>	/k/	成套工具
<b>	/乙/	少量
<d>	/d/	喧嚣
<j>	/dʒ/	喜悦
<f>	/F/	乐趣
<v>	/v/	货车
<z>	/z/	压缩
<l>	/升/	唇
<米>	/米/	头脑
<w>	/w/	湿

也有具有多个音位（或语音）值的专着：

There are also monographs with more than one phonemic (or phonetic) value:

(13)

字素	音素/音位变体	例子
<y>	/j/	年
<n>	/n/, [m], [ɱ], [ŋ]	ip, input, inform, 墨水
<s>	/s/, /z/, /ʃ/, /ʒ/	啜饮，玫瑰，强迫症，迷惑
<g>	/g/, /dʒ/	得到了，杜松子酒
<x>	/ks/, /gz/, /kʃ/, /gʒ/	性爱，确切，奢华，奢华

对于字素 <y>，我们将诸如年份之类的词视为以辅音开头：上颚近似词 /j/。[j] 和 [w] 这两个音通常被称为半辅音，因为它们的发音类似于元音，但它们在音节中占据辅音位置，即声母位置。

For the grapheme , we are treating words such as year as beginning with a consonant: the palatal approximant /j/. The sounds [j] and [w] are often referred to as semi-consonants, since they are vowel-like in their articulation, but they occupy a consonantal position in syllables, namely the onset position.

就字素 <n> 而言，我们已经看到音素 /n/ 经历了鼻同化：如果后跟双唇辅音，则实现是双唇的，如input；如果后跟唇齿辅音，则它的实现是唇齿，如inform；如果后跟软腭辅音，则它的实现是软肋，如ink中。这些双唇、唇齿和软腭实现都是 /n/ 音素的变位音。

As far as the grapheme is concerned, we have seen that the phoneme /n/ undergoes nasal assimilation: if followed by a bilabial consonant, the realization is bilabial, as in input; if followed by a labio-dental consonant, its realization is labio-dental, as in inform; and if followed by a velar consonant, its realization is velar, as in ink. These bilabial, labio-dental and velar realizations are all allophones of the /n/ phoneme.

由于历史原因，<s> 字素对应于四个音素：腭齿龈发音 /ʃ/ 和 /ʒ/ 出现是因为以下辅音是腭音近似值（非正式地称为yod）：序列 [s] + [ j] 导致了一个被称为聚结的同化过程，其中一系列的牙槽和腭导致了腭-牙槽的发音。在连接语音中可以看到相同的过程：词尾 /s/ 后跟 yod 可以导致同化。这可以在滚石乐队的一首名为“Miss You”的歌曲中听到，米克·贾格尔 (Mick Jagger) 发音为 [ˈmɪʃə]。类似地，词尾 /z/ 后跟词首 /ε/ 可以产生 [ʒ]，如He's yours，发音为 [hiˈʒɜ:z]。玫瑰之类的字, rise的过去时，最初有一个清音 [s]，通常带有词尾 schwa：[ro:sə]。清音段通常会在元音间（元音之间）变成浊音：由于元音通常是浊音，因此会发生同化过程，导致元音间的声音变浊。因此 [ro:zə]，后来失去了词尾 schwa。很久以后，/o:/ 在 RP 中双元音化，产生了现在的 [ɪəʊz] 发音。

The grapheme corresponds to four phonemes for historical reasons: the palato-alveolar pronunciations /ʃ/ and /ʒ/ came about because the following consonant was a palatal approximant (known informally as a yod): the sequence [s] + [j] led to an assimilation process known as coalescence, in which a sequence of an alveolar and a palatal results in a palato-alveolar articulation. The same process can be seen in connected speech: a word-final /s/, when followed by a yod, can lead to assimilation. This can be heard in a song by the Rolling Stones, entitled ‘Miss You’, pronounced [ˈmɪʃə] by Mick Jagger. Similarly, a word-final /z/ followed by a word-initial /ε/ can yield a [ʒ], as in He’s yours, pronounced [hiˈʒɜ:z]. Words such as rose, the past tense of rise, initially had a voiceless [s], often with a word-final schwa: [ro:sə]. It is common for voiceless segments to become voiced intervocalically (between vowels): since vowels are typically voiced, an assimilation process takes place which results in the intervocalic sound undergoing voicing. Thus [ro:zə], which later lost the word-final schwa. Much later, the /o:/ diphthongized in RP, resulting in the present-day [ɪəʊz] pronunciation.

<g> 字素对应于 /g/ 音素，但也对应于 /dʒ/ 音素，同样是出于与同化有关的历史原因：由于gin、gibberish和gist等词中的元音是高前元音，因此historical /g/ 经历了腭化，变成了腭-牙槽音。在gibe和Giles等词中，元音是中古英语中的高位元音 /i:/，因此引发了相同的腭化过程。不幸的是，并不是所有的 /g/ + 前高元音序列都经历了这个过程，因此像give和gig这样的词虽然包含前高元音，却保留了软喉发音 /g/。一些词，例如专有名称Gill, 可以任意发音，女性名字Gill的发音为 [dʒ]，男性名字Gill的发音为 [g]。类似地，gill这个词在表示鱼的一部分时，发音为软腭塞音，而称为 gill 的测量单位则发音为腭牙槽值。

The grapheme corresponds to the /g/ phoneme, but also corresponds to the /dʒ/ phoneme, again for historical reasons relating to assimilation: since the vowel in words such as gin, gibberish and gist is a high front vowel, the historical /g/ underwent palatalization, and became palato-alveolar. In words such as gibe and Giles, the vowel was the high-front /i:/ in Middle English, and thus induced the same palatalization process. Unfortunately, not all /g/ + high-front vowel sequences underwent this process, so that words such as give and gig, while containing a high front vowel, have retained the velar pronunciation /g/. Some words, such as the proper name Gill, can be pronounced either way, with the female name Gill being pronounced with a [dʒ], and the male name Gill being pronounced with a [g]. Similarly, the word gill, when meaning a part of a fish, is pronounced with a velar stop, whereas the unit of measurement known as a gill is pronounced with the palato-alveolar value.

<x> 字素对应于两个音素的序列。如果我们将 /ks/ 序列作为“基本”值，我们可以看到发声和腭化的相同历史过程已经发生以产生其他三个值。exact等单词中的 /gz/ 序列是从元音间浊音同化产生的。在像luxury这样的词中，序列 /ksj/ 引起 /sj/ 序列的合并，导致 /kʃ/。在像luxurious这样的词中，合并和元音间的浊化都导致了 /kʒ/ 序列。压力也在这里发挥作用：如果重读前面的元音，如sex和luxury，我们有清音值；如果重读以下元音，如exact和luxury，我们有一个浊音值。

The grapheme corresponds to sequences of two phonemes. If we take the /ks/ sequence to be the ‘basic’ value, we can see that the same historical processes of voicing and palatalization have taken place to yield the other three values. The /gz/ sequences in words like exact have arisen from intervocalic voicing assimilation. In words like luxury, the sequence /ksj/ gave rise to coalescence of the /sj/ sequence, resulting in /kʃ/. In words like luxurious, both coalescence and inter-vocalic voicing have resulted in the /kʒ/ sequence. Stress also plays a role here: if the preceding vowel is stressed, as in sex and luxury, we have a voiceless value; if the following vowel is stressed, as in exact and luxurious, we have a voiced value.

11.3.2 辅音二合字母和三合字母

11.3.2 Consonant Digraphs and Trigraphs

有具有单个音位值的辅音二合字母和三合字母：

There are consonant digraphs and trigraphs with a single phonemic value:

(14)

字素	音素	例子
<ck>	/k/	钟
<ph>	/F/	照片
<嘘>	/ʃ/	船
<右手>	/ɹ/	大黄
<dg>	/dʒ/	边缘
<d¡>	/dʒ/	附件
<吴>	/ŋ/	唱歌
<t>	/tʃ/	痒

除了多值的辅音专音外，还有多值的二合音和三合音。这些都是：

In addition to the consonant monographs with more than one value, there are consonant digraphs and trigraphs with more than one value. These are:

(15)

字素	音素	例子
<顾>	/g/, /gw/	守卫，痛苦
<曲>	/k/, /kw/	独特的，女王
<ch>	/tʃ/, /k/, /ʃ/	椅子，混乱，别致
<gh>	/g/, /f/, 零	食尸鬼，坚韧，大腿
<日>	/θ/, /ð/	想，这
<施>	/sk/, /ʃ/	学校，伤感
<sc>	/s/, /ʃ/, /sk/	场景,意识,冲刷
<蛋>	/g/, /dʒ/	鸡蛋，夸张
<抄送>	/k/, /ks/	帐户，口音
<ss>	/s/, /ʃ/	接吻，任务

<gu> 二合字母主要对应于 /g/ 音素，如在guide和vague中，但也可以对应于 /gw/ 序列。这只能在词的中间找到，那里没有语素边界，如在anguish、language和penguin中，但在例如vaguely中则没有。

The digraph mostly corresponds to the /g/ phoneme, as in guide and vague, but can correspond to the /gw/ sequence. This can be found only in the middle of words, where there is no morpheme boundary, as in anguish, language and penguin, but not in, for instance, vaguely.

像queen这样的词中的 <qu> 二合字母是在诺曼征服之后由法国抄写员引入的：像这样的词以前是用 <cw> 序列写的。诸如unique之类的词是从法语中借来的，带有法语拼写。

The digraph in words like queen was introduced by French scribes after the Norman Conquest: words like this had previously been written with a sequence. Words such as unique have been borrowed from French with their French spellings.

<ch> 二合字母取代了单词中的 <c> 专论，例如church，以前发音为 /k/，现在发音为 /tʃ/。值 /k/ 可以在许多源自希腊语的词中找到，例如chaos、chemist和psychology。/ʃ/ 值出现在最近的法语外来词中，例如别致、香槟和小册子。我们没有列出值 /x/（清软腭擦音），因为它不会出现在大多数英语变体中。然而，它确实出现在苏格兰英语中，例如loch (/lɔx/)，而不是lock (/lɔk/)。虽然一些非苏格兰人发音诸如loch之类的单词时带有 [x]，但大多数人在尝试苏格兰单词（例如地名Auchtermuchty）、德语单词（例如Bach ）或西班牙语单词（例如rioja ）时会发出 [k] 。

The digraph replaced the monograph in words like church, which had previously been pronounced with a /k/, and is now pronounced with a /tʃ/. The value /k/ can be found in many words of Greek origin, such as chaos, chemist and psychology. The /ʃ/ value is found in fairly recent loanwords from French, such as chic, champagne and brochure. We have not listed the value /x/ (the voiceless velar fricative), since it does not occur in most varieties of English. It does, however, occur in Scottish English, in words such as loch (/lɔx/), as opposed to lock (/lɔk/). While some non-Scottish speakers pronounce words such as loch with a [x], most produce a [k] when attempting Scottish words such as the place-name Auchtermuchty, German words such as Bach or Spanish words such as rioja.

<gh> 二合字母用于表示清软腭擦音 /x/，可以在西班牙语单词中听到，例如ajo（“大蒜”）或德语单词，例如Buch（“书”）。清软腭擦音不再存在于 RP 或 GA 中：它在许多词中被省略，例如thigh、daughter和thought，这就是我们将其值表示为零的原因。清软颚擦音在年长的苏格兰人中仍然可以听到，他们将daughter发音为 [dɔxtɪɾ] 并认为是 [θɔxt]。

The digraph used to represent the voiceless velar fricative /x/, which can be heard in Spanish words like ajo (‘garlic’) or German words like Buch (‘book’). The voiceless velar fricative no longer exists in RP or GA: it was elided in many words, like thigh, daughter and thought, which is why we have represented its value as zero. The voiceless velar fricative can still be heard among older speakers of Scots, who pronounce daughter as [dɔxtɪɾ] and thought as [θɔxt].

大多数拼写为 <th> 的单词对应于 /θ/，因为包含 /ð/ 的 <th> 单词是非词汇类别的单词（功能词），例如the、this、that、these、those和then。虽然这些都是高频词，但数量很少，这是典型的虚词。

Most words spelled with correspond to /θ/, since words which contain /ð/ are words of a non-lexical category (function words), such as the, this, that, these, those and then. While these are high-frequency words, there are very few of them, which is typical of function words.

用 <sch> 三字母拼写的单词，例如schmaltzy、schmuck和schmooze通常源自意第绪语。意第绪语是德系犹太人使用的一种日耳曼语（词汇来自希伯来语、阿拉姆语和斯拉夫语）。进入英语的意第绪语词大多来自纽约的犹太社区。其中一些词，例如schlep（“携带重物”或“在某处缓慢旅行”），在整个美国都不为人所知，过去在英式英语中也不使用，但schlep开始出现在英国报纸上。其他 <sch> 词是来自德语的外来词，例如schnapps（一种德国饮料）和schnitzel（用面包屑炸的肉）。school等词是从拉丁语借来的，其中 <sch> 序列对应于 /sk/，并且在英语中保留了该序列一千多年。

Words spelled with the trigraph, such as schmaltzy, schmuck and schmooze are often of Yiddish origin. Yiddish is a Germanic language (with vocabulary from Hebrew, Aramaic and Slavic languages) spoken by Ashkenazi Jews. The Yiddish words which have found their way into English mostly come from New York’s Jewish community. Some of these words, such as schlep (‘to carry around a heavy object’ or ‘to travel somewhere slowly’), are not known throughout the United States, and used not to be used in British English, but schlep is beginning to appear in British newspapers. Other words are loanwords from German, such as schnapps (a German drink) and schnitzel (meat fried in breadcrumbs). Words such as school were borrowed from Latin, in which the sequence corresponded to /sk/, and has retained that sequence in English for over a thousand years.

许多用 <sc> 拼写的词是从法语借来的。他们最初在拉丁语中有 /sk/ 序列，但在他们被借用时 /k/ 在法语发音中被省略了。

Many words spelled with were borrowed from French. They initially had /sk/ sequences in Latin, but the /k/ had been elided in the French pronunciation by the time they were borrowed.

序列 <gg> 和 <cc> 对应于与 <g> 和 <c> 相同的值。英语中的双字不对应于长辅音音素（通常称为 geminates），这与意大利语不同，在意大利语中，mamma和pizza等拼写对应于发音 /ˈmam:a/ 和 /ˈpit:sa/，其中冒号表示长度。一些相同辅音字素的序列对应于跨语素边界的两个相同辅音音素的序列，如unnerve，其形式为 /ˌʌn + ˈnɜ:v/，实现为 [ˌʌnˈnɜ:v]。英语中两个相同音素的序列与单个长（双）音素不同：英语中没有 /n:/ 这样的音素。pizza with a geminate consonant 反映了英语没有双辅音音素的事实。

The sequences and correspond to the same values as and . Doubled graphemes in English do not correspond to long consonant phonemes (often called geminates), unlike in Italian, where spellings such as mamma and pizza correspond to the pronunciations /ˈmam:a/ and /ˈpit:sa/, where the colon indicates length. Some sequences of identical consonant graphemes correspond to sequences of two identical consonant phonemes across a morpheme boundary, as in unnerve, which has the form /ˌʌn + ˈnɜ:v/, realized as [ˌʌnˈnɜ:v]. A sequence of two identical phonemes in English is not the same thing as a single, long (geminate) phoneme: there is no such phoneme as /n:/ in English. The fact that English speakers do not pronounce Italian loanwords such as pizza with a geminate consonant reflects the fact that English has no geminate consonant phonemes.

<ss> 序列将 /s/ 作为其“基本”值，如kiss和massive，但我们上面讨论的腭化过程的种类在历史上有过 yod 的词中产生了 /ʃ/ ，如mission和assure。

The sequence has /s/ as its ‘basic’ value, as in kiss and massive, but the kinds of palatalization process we discussed above have yielded /ʃ/ in words which have had a yod historically, like mission and assure.

11.3.3 不发音的辅音字素

11.3.3 Unpronounced Consonant Graphemes

随着英语的发展，一些词首辅音已被省略。这些包括 /k/ 和 /g/ 在带有 /n/ 的分支首字母中，因此以下二合字母不再对应于首字母音素序列：

Some word-initial consonants have been elided as English has evolved. These include /k/ and /g/ in branching onsets with /n/, so that the following digraphs no longer correspond to sequences of onset phonemes:

(16)

字素	音素	例子
<kn>	/n/	知道
<gn>	/n/	侏儒

其他词是从希腊语借来的，其语音限制允许诸如 /ps/ 和 /pt/ 之类的起始序列。由于英语音位学不允许出现这样的序列，因此出现了以下对应关系：

Other words were borrowed from Greek, whose phonotactic constraints allow onset sequences such as /ps/ and /pt/. Since English phonotactics do not allow such sequences in onsets, the following correspondences have arisen:

(17)

字素	音素	例子
<附言>	/秒/	精神
<点>	/吨/	托勒密

反映省略历史过程的其他字形序列包括：

Other graphemic sequences which reflect historical process of elision include the following:

(18)

字素	音素	例子
<写>	/ɹ/	写
<wh>	/w/, /h/	发牢骚，整个

古英语在whole之类的单词中有一个起始序列 /hw/ 。在 RP 和 GA 中，/h/ 已被删除，而拼写仍然表示两个音素的簇。在 SSE 和某些美式英语变体中，/hw/ 序列变成了 [ʍ]，一种同时具有双唇和软腭发音的清擦音，很像吹灭蜡烛时发出的声音。which和witch等对在 SSE 中是最小对：/ʍɪtʃ/ vs /wɪtʃ/，而在 RP 和 GA 中它们是同音字。

Old English had an onset sequence /hw/ in words like whole. In RP and GA, the /h/ has since elided, while the spelling still indicates a cluster of two phonemes. In SSE, and in some varieties of American English, the /hw/ sequence has become [ʍ], a voiceless fricative which has both a bilabial and a velar articulation, rather like the sound one makes when blowing out a candle. Pairs such as which and witch are minimal pairs in SSE: /ʍɪtʃ/ vs /wɪtʃ/, whereas they are homophones in RP and GA.

也有对应于单个音素的词尾二合字母，因为词尾辅音在语言的演变过程中已经消失：

There are also word-final digraphs which correspond to a single phoneme, since the word-final consonant has elided during the evolution of the language:

(19)

字素	音素	例子
<mn>	/米/	圣歌
<mb>	/米/	梳子
<gn>	/n/	符号
<通用>	/米/	范例

在形态相关的词中，词根-词尾辅音不再是词尾，因此经常被保留，如下所示：

In morphologically related words, the root-final consonant is no longer word-final, and has therefore often been retained, as in the following:

(20)

拼写	发音
赞美诗	[ˈhɪmnəə]
签字人	[ˈsɪgnətɹi]
典范的	[ˌpæɹədɪgˈmætɪk]

在词中位置，<t> 通常不在以下序列中发音：<st> 后跟元音字母，如christen和 <stl>，如castle。

In word-medial position, is often not pronounced in the following sequences: followed by a vowel grapheme, as in christen, and , as in castle.

11.3.4 音位学和对比音位学

11.3.4 Graphophonemics and Contrastive Phonemics

一些英语作为外语的学习者遇到的问题既涉及英语的字形对应关系，也涉及其母语与英语音位系统之间的差异，尤其是 RP 英语。一个例子涉及中元音。通常会发现具有如下核心单音字系统的语言：

Some learners of English as a foreign language encounter problems which relate both to the English graphophonemic correspondences and to the difference between the phoneme system of their native language and that of English, especially RP English. An example concerns mid vowels. It is common to find languages with a core monoph-thong system like the following:

(21)

/一世/		/你/
/e/		/o/
/ε/		/ɔ/
	/ɐ/

核心单元音系统是指忽略例如法语音素 /ε̃/、/ɑ̃/ 和 /ɔ̃/ 等鼻音化单元音或法语音素 /y/、/ø/ 和前圆唇元音的系统/œ/，所有这些都被语言学家称为“标记”，即在世界语言中相对不常见。

By core monophthong system is meant a system ignoring, for instance, nasalized monophthongs such as the French phonemes /ε̃/, /ɑ̃/ and /ɔ̃/, or front rounded vowels, such as the French phonemes /y/, /ø/ and /œ/, all of which are said by linguists to be ‘marked’, i.e. relatively uncommon in the world’s languages.

在上图中，有两个高元音（/i/和/u/），两个中高元音（/e/和/o/），两个中低元音（/ε/和/ɔ/），加上一个既不是全前也不是全后的单个低元音，这就是我们选择 /ɐ/ 表示的原因。具有这种系统的语言的例子是罗曼语系的法语、西班牙语和意大利语。如果说这种核心系统的语言，RP的元音音素系统是一个很大的挑战。首先，它有两个高后圆唇元音（/u:/ 和 /ʊ/），而不是一个，这是世界语言的标准。它还有两个“a”型元音（/æ/ 和 /ɑ:/），甚至三个，如果我们认为元音 /ʌ/ 更像是一个 'a' 的声音，因此在缺乏 /æ/–/ʌ/–/ɑ:/ 对比的语言中经常用 [ɐ] 发音。RP 的后中元音特别困难，因为 RP 有三个“o”型音素：/əʊ/（在中古英语中曾经是单元音 /o:/）、长/紧 /ɔ:/ 和短/松/ɑ/. 对于母语只有两个“o”型音素的人来说，一个语音难题就出现了：如何将两个感知类别扩展为三个。

In the above diagram, there are two high vowels (/i/ and /u/), two high-mid vowels (/e/ and /o/), two low-mid vowels (/ε/ and /ɔ/), plus a single low vowel which is neither fully front nor fully back, which is why we have chosen the /ɐ/ representation. Examples of languages with such a system are the Romance languages French, Spanish and Italian. If one speaks a language with this kind of core system, the vowel phoneme system of RP is quite a challenge. To begin with, it has two high back rounded vowels (/u:/ and /ʊ/), rather than one, which is the norm in the world’s languages. Its also has two ‘a’-type vowels (/æ/ and /ɑ:/), or even three, if we consider that the vowel /ʌ/ is rather like an ‘a’ sound, and thus often pronounced with an [ɐ] in languages which lack an /æ/–/ʌ/–/ɑ:/ contrast. The back mid vowels of RP are especially difficult, since RP has three ‘o’-type phonemes: /əʊ/ (which used to be monophthongal /o:/ in Middle English), long/tense /ɔ:/ and short/lax /ɑ/. For speakers whose native language has only two ‘o’-type phonemes, a phonological difficulty arises: how to expand the two perceptual categories into three.

我们头脑中的母语音位系统是一个感知类别系统，我们通过它来解码语音。很难如果外语中的三向对比对应于我们母语中的双向对比，那么我们会感知到外语中的三向对比。例如，英语在齿龈和腭齿龈擦音之间有两种对比：/s/–/ʃ/ 和 /z/–/ʒ/。如果说英语的人试图学习波兰语，那么摩擦音就会出现困难，因为波兰语在齿龈、腭-齿龈和腭前擦音之间存在三向对比，转录为 /s/、/ʃ/ 和 /ś/。因为我们说英语的人习惯于在口腔的这个区域只感知两类清摩擦音，所以我们很难感知和产生三向波兰语对比。

The phoneme system of our native language, which we have in our heads, is a system of perceptual categories, via which we decode speech. It is difficult to perceive a three-way contrast in a foreign language if it corresponds to a two-way contrast in our native language. For instance, English has a two-way contrast between alveolar and palato-alveolar fricatives: /s/–/ʃ/ and /z/–/ʒ/. If an English speaker tries to learn Polish, a difficulty arises with the fricatives, since Polish has a three- way contrast between alveolar, palato-alveolar and pre-palatal fricatives, transcribed as /s/, /ʃ/ and /ś/. Because we English speakers are so accustomed to perceiving only two categories of voiceless fricative in this region of the oral cavity, we have difficulty with both perceiving and producing the three-way Polish contrast.

对于正在学习英语的多种语言的使用者来说，同样的困难适用于 /əʊ/、/ɔ:/ 和 /ɑ/ 之间的 RP 对比。标准的苏格兰英语对这些说话者来说会更容易，因为它的核心单元音系统与上面描述的系统非常相似：只有一个“u”型声音，双向 /e/–/ε/ 区分，双向 /o /–/ɔ/ 区分并且只有一个 'a' 音素（尽管 SSE 确实有 /ʌ/ 音素）。“o”音的语音困难与变化无常的英语拼写结合在一起，因为正如我们所见，三种不同的“o”音有多种拼写方式。我们已经看到 <o> 字素可以对应于 /əʊ/、/ɔ:/ 或 /ɑ/。然而，对于非母语人士学习 RP 来说，隧道尽头有一些曙光：我们在上面引用了字音规律，这些规律决定了，在很大程度上，我们找到了哪些值。此外，序列 <aught>、<ought> 和 <aw> 系统地对应于 RP 长元音 /ɔ:/，如caught, seeded and lawn，而不是 RP /əʊ/。对于英语学习者来说，这一点值得牢记，因为存在 RP 最小对，例如loan和lawn，它们分别具有 /əʊ/ 和 /ɔ:/。有一则轶事或许可以说明这个问题：当我们第一次在法国买房子时，一位法国同事问了我一个问题。我以为问题是“你有贷款吗？”，我回答说“是的，当然：我们没有足够的钱支付现金买房子。” 他的意思是“你有草坪吗？”，但他为lawn中的 <aw> 序列产生了 /əʊ/ ，而不是 /ɔ:/ 发音。

For speakers of many languages who are learning English, the same difficulty applies to the RP contrast between /əʊ/, /ɔ:/ and /ɑ/. Standard Scottish English would be easier for such speakers, since its core monophthong system is very similar to the one depicted above: only one ‘u’-type sound, a two-way /e/–/ε/ distinction, a twoway /o/–/ɔ/ distinction and only one ‘a’ phoneme (though SSE does have the /ʌ/ phoneme). The phonological difficulty with ‘o’ sounds is compounded by the vagaries of English spelling since, as we have seen, there are various ways of spelling the three different ‘o’ sounds. We have seen that the grapheme can correspond to either /əʊ/, /ɔ:/ or /ɑ/. However, there is some light at the end of the tunnel for non-native speakers learning RP: we have cited graphophonemic regularities above which determine, to a large extent, which values we find. In addition, the sequences , and systematically correspond to the RP long vowel /ɔ:/, as in caught, sought and lawn, and not to RP /əʊ/. This is a point worth bearing in mind for learners of English, since there are RP minimal pairs such as loan and lawn, which have, respectively, /əʊ/ and /ɔ:/. An anecdote might serve to highlight the problem: when we first bought a house in France, a French colleague asked me a question about it. I thought the question was ‘Do you have a loan?’, to which I answered ‘Yes, of course: we’re not wealthy enough to have paid cash for the house.’ What he meant was ‘Do you have a lawn?’, but he produced /əʊ/ for the sequence in lawn, rather than the /ɔ:/ pronunciation.

总结英语拼写与发音的对应关系：我们上面给出的规律表明这些并不完全混乱。但是，我们提供的“规则”也有许多例外情况，这些例外情况使英语字音学比原本可能的情况更加混乱。例如，我们已经看到像panel这样的词真的应该写成pannel，就像在channel中一样，带有双辅音字素，而我自己的名字Philip应该写成Phillip, 表示前重读元音的检查值 ([ɪ])。然而，也许最好是强调规律性，而不是列出一长串例外情况：毕竟，除非有规则形式，否则不可能有不规则形式。

To conclude on English spelling-to-pronunciation correspondences: the regularities we have given above show that these are not completely chaotic. There are, though, many exceptions to the ‘rules’ we have provided, and those make English graphophonemics rather more messy than it might otherwise have been. We have seen, for instance, that words such as panel really ought to be written pannel, as in channel, with a double consonant grapheme, and my own name, Philip, ought to be written Phillip, to indicate the checked value ([ɪ]) of the pre-final stressed vowel. However, it is best, perhaps, to emphasize the regularities, rather than make a very long list of exceptions: one cannot, after all, have irregular forms unless there are regular forms.

练习

Exercises

gueqrecard_001

1在www.wiley.com/go/carrphonetics上收听音轨 11.1，由 RP 扬声器制作。对于录音中的每个词（如下所列），说出是否元音对应于元音字素的自由值或勾选值。解释为什么。在斜括号中为每个元音音素提供符号。

（亚当

(b) 女士

(d) 定义

(e) 蒂姆

(f) 时间

(g) 点

(h) 溺爱

(i) 邓

(j) 沙丘

gueqrecard_001

2 聆听由 RP 扬声器制作的音轨 11.2 。对于录音中每个单词中的重读音节，说出元音字素对应于自由值还是检查值，并解释拼写如何编码这些值。

（一）锥度

(b) 攻丝机

（三）彼得

(d) 彼得

(e) 消瘦

(f) 固定

(g) 溺爱

(h) 打点

(i) 更敏锐

(j) 口吃

gueqrecard_002

3 聆听RP 扬声器制作的音轨 11.3。对于以下每个单词中的重读音节，说出元音字素对应于自由值还是检查值，并解释这与单词拼写方式的关系。

(a) 理智

(b) 插曲

(d) 歌剧

gueqrecard_002

4 聆听由 RP 扬声器制作的音轨 11.4 。<gh> 二合字母在以下每个单词中对应哪个值？

（粗略

(二) 通过

12 英语口音的变化

Variation in English Accents

12.1 简介

12.1 Introduction

在本章中，我们将考虑口音变化的一些一般方面。第 13 章简要概述了几种英语口音：伦敦英语、泰恩赛德英语、标准苏格兰英语 (SSE)、纽约市英语、德克萨斯英语、澳大利亚英语和印度英语，然后概述了各种口音随着时间的推移导致口音分歧的现象。

In this chapter, we will consider some general aspects of accent variation. In chapter 13, a brief overview is given of several accents of English: London English, Tyneside English, Standard Scottish English (SSE), New York City English, Texan English, Australian English and Indian English, followed by an outline of the sorts of phenomena which give rise to divergence of accents over time.

我们在本书中提到的三种口音（GA、RP 和 SSE）在社会上被视为“标准”口音。“标准”这个概念是一个社会概念：没有语言学家会声称存在任何内在语音或语音优势的连贯概念，因为这样的概念根本没有任何语音或音位意义。毫无疑问，许多人认为某些口音优于其他口音，或者将某些口音视为标准口音而将其他口音视为非标准口音。但这些判断是建立在非语言因素的基础上的，与相关社会的社会态度有关。从严格的语言学角度来看，这样的判断显然完全是武断的。例如，英国的标准口音 RP 是非风音的，因此，一些（也许很多）英国人认为 RP 的无声口音比英格兰西部许多地区的声带口音更有声望。但是标准口音 SSE 和 GA 是rhotic，而在美国，通常被判断为比非rhotic 美国口音更有声望的是rhotic 口音（这种判断不会出现在苏格兰，所有本土口音都是rhotic ).

Three of the accents we have referred to in this book (GA, RP and SSE) are viewed socially as ‘standard’ accents. The notion ‘standard’ is a social one: no linguist would claim that there is any coherent notion of inherent phonetic or phonological superiority, since such a notion simply does not make any phonetic or phonological sense. There can be no doubt that many people judge some accents to be superior to others, or take some accents to be standard accents and others to be non-standard accents. But those judgements are founded on non-linguistic factors, to do with social attitudes in the societies in question. From a strictly linguistic point of view, such judgements are, quite clearly, entirely arbitrary. For example, RP, the standard accent in England, is non-rhotic, and the non-rhoticity of RP is therefore judged by some (perhaps many) English people to be more prestigious than the rhotic accents found in many of the Western parts of England. But the standard accents SSE and GA are rhotic, and in the United States, it is the rhotic accents which are often judged to be more prestigious than the non-rhotic American accents (the judgement cannot arise in Scotland, where all native accents are rhotic).

显然，是社会态度决定了对口音的这种判断，而不是口音本身的语音和音位特性。社会判断通常认为某些口音比其他口音“更丑陋”或“更刺耳”。就语音学和音位学而言，这些判断也完全是任意的。例如，如果单词say在伦敦英语中发音为 [saɪ]，被说普通话的人判断为“丑陋”，那么说普通话的人对sigh一词的自己的发音，作为 [saɪ]，也应该将 RP 扬声器视为“丑陋”。诸如此类的案例表明，根本不可能是声音 [saɪ] 本身的任何语音属性导致了审美判断。相反，这样的判断也源于并反映了社会对什么可以更广泛地称为“生活方式”的态度。在英国，大多数被判断为“丑陋”或“不文明”的非标准口音是在工业或后工业城市地区使用的；经常被引用的例子是伦敦、伯明翰、利物浦、贝尔法斯特、格拉斯哥和泰恩赛德的工人阶级口音。美国也有类似的判断，例如，通常被称为“布鲁克林口音”的纽约市口音（尽管它并不局限于纽约市的布鲁克林区）。

Clearly, it is social attitudes which determine such judgements about accents, rather than the phonetic and phonological properties of the accents themselves. It is common to find social judgements to the effect that some accents are ‘uglier’ or ‘harsher’ than others. These judgements too are entirely arbitrary as far as phonetics and phonology are concerned. For instance, if the word say, pronounced as [saɪ] in London English, is judged ‘ugly’ by an RP speaker, then the RP speaker’s own pronunciation of the word sigh, as [saɪ], ought also to strike the RP speaker as ‘ugly’. Cases such as this show that it simply cannot be any phonetic properties of the sounds [saɪ] in and of themselves which induce the aesthetic judgement. Rather, such judgements also derive from and reflect social attitudes about what might, rather broadly, be called ‘ways of life’. In Britain, most non-standard accents which are judged ‘ugly’ or ‘uncivilized’ are spoken in industrial or post-industrial urban areas; examples often cited are the working-class accents of London, Birmingham Liverpool, Belfast, Glasgow and Tyneside. Similar sorts of judgement are made in the United States, with respect to, for example, the broad New York City accent often referred to as ‘the Brooklyn accent’ (though it is not confined to the Brooklyn district of New York City).

相比之下，不标准的乡村口音通常被认为是“古怪的”，而不是“丑陋的”；例如来自苏格兰高地、英格兰西部乡村或美国南部各州的口音。这些审美判断很可能源于口音与说话者真实或想象的生活方式有意识或无意识的联系。因此，我们将从这种社会态度中退后一步，并尝试从语音学家和音韵学家的角度来研究口音的变化。^1个

Non-standard rural accents are, by contrast, often judged ‘quaint’, rather than ‘ugly’; examples are accents from the Highlands of Scotland, from the West Country in England, or from the US Southern states. It is highly likely that these aesthetic judgements arise from the conscious or unconscious association of accents with the real or imagined ways of life of those who speak them. We will therefore stand back from such social attitudes and attempt to examine accent variation from the point of view of the phonetician and the phonologist.1

12.2 口音之间的系统差异与实现差异

12.2 Systemic vs Realizational Differences between Accents

让我们首先考虑英格兰北部的许多口音与英格兰南部的许多口音之间的差异之一。在后一种口音中，/ʊ/ 和 /ʌ/ 之间存在语音对比，可以成对观察到，例如book/buck、rook/ruck、put/putt等等。许多北方英语口音都没有这种区别，²每个口音中都有 /ʊ/。也就是说，许多北方英语口音只是缺少 /ʌ/ 音素，因此没有 /ʊ/ 与 /ʌ/ 的区别：对他们来说，put和putt等对是同音词，而不是最小对。我们将这种差异称为系统性差异³两种口音之间：说话者的一组语音对比（具体来说，在这种情况下是元音对比）不同。

Let us begin by considering one of the differences between many accents in the North of England and many of those spoken in the South of England. In the latter accents, there is a phonological contrast between /ʊ/ and /ʌ/, which can be observed in pairs such as book/buck, rook/ruck, put/putt and many others. That distinction is missing in many Northern English accents,2 which have /ʊ/ in each member of the pair. That is, many Northern English accents simply lack the /ʌ/ phoneme, and thus the /ʊ/ vs /ʌ/ distinction: for them, pairs such as put and putt are homophones, not minimal pairs. We will refer to this sort of difference as a systemic difference3 between two accents: the set of phonological contrasts (specifically, in this case, the vowel contrasts) of the speakers differ.

系统性差异已得到广泛证实。例如，在许多苏格兰英语和苏格兰口音中，没有等同于 /æ/ 与 /ɑ:/ 区别，在ant/aunt、palm/Pam等 RP 最小对中发现的那种区别。在苏格兰英语中，这样的对中的每个成员都包含相同的元音音素 /ɐ/，实现为 [ɐ]。同样，在非苏格兰口音中算作最小对的对在苏格兰英语中是同音的。类似地，正如我们所见，虽然 RP 在 [ɒ]、[ɑ:] 和 [ɔ:] 之间有三向区别，但 GA 的许多说话者在 [ɔ:] 和 /ɑ/ 之间只有两种区别. 这种系统性差异意味着通过这些音素区分的词组有所不同，如下所示：

Systemic differences are widely attested. In many Scottish English and Scots accents, for instance, there is no equivalent of the /æ/ vs /ɑ:/ distinction, of the sort found in RP minimal pairs such as ant/aunt, palm/Pam, etc. In Scottish English, each member of such pairs contains the same vowel phoneme, /ɐ/, which is realized as [ɐ]. Again, pairs which count as minimal pairs in non-Scottish accents are homophonous in Scottish English. Similarly, as we have seen, while RP has a three-way distinction between [ɒ], [ɑ:] and [ɔ:], many speakers of GA have only a two-way distinction between [ɔ:] and /ɑ/. This systemic difference means that there is a difference in the sets of words which are distinguished by means of these phonemes, as follows:

(1)

类型的话	回波	遗传算法
棕榈	/ɑ:/	/ɑ/
抓住	/ɔ:/	/ɔ:/ 或 /ɑ/
婴儿床	/ɒ/	/ɑ/
咖啡	/ɒ/	/ɔ:/

系统差异不仅限于元音系统。辅音系统系统性差异的一个例子是在许多苏格兰口音中发现的 /ʍ/ 和 /w/ 之间的对比，这种对比出现在最小的对中，例如whales与Wales、whin与win、what与watt等。大多数非苏格兰口音都没有这种对比，它们有 /w/ 音位，但没有 /ʍ/ 音位。在这些口音中，上述对是同音字，而不是最小对。

Systemic differences are not restricted to vowel systems. An example of a systemic difference in consonant systems is the contrast, found in many Scottish accents, between /ʍ/ and /w/, which is found in minimal pairs such as whales vs Wales, whin vs win, what vs watt, etc. That contrast is absent in most non-Scottish accents, which have a /w/ phoneme, but not a /ʍ/ phoneme. In those accents, the above pairs are homophones, rather than minimal pairs.

重音之间的差异并不等于语音对比系统中的差异。考虑我们在第 7 章中对暗色和非暗色 /l/ 的讨论：我们在那儿 (7.6) 说过 RP 中有一个同位异音规则，大意是 /l/ 被实现为软化（“暗”）侧音，当它出现在押韵中，但当它出现在词首时被实现为非软化（“非黑暗”）侧音。在苏格兰南部的许多口音、澳大利亚英语、GA 和英格兰北部的一些口音中，/l/ 在所有位置都被实现为暗侧音，因此 /lʌl/ ( lull )，例如，实现为 [ɫʌɫ]，而不是 [lʌɫ]。另一方面，在泰恩赛德英语中，/l/ 在所有位置上始终被理解为清晰的侧音，因此 /lʌl/ ( lull )⁴实现为 [l^j ʊl^j]. 毫无疑问，在这些情况下，底层对比系统存在差异：所有三种重音在 /l/ 和其他辅音音素（例如 /r/）之间都有对比，但方式有所不同/l/ 实现了。虽然在某些语言中，清晰和黑暗的侧音之间的区别是对比的，但在任何这些英语口音中都不可能假设这种语音对比：在苏格兰南部和澳大利亚的情况下，只有黑暗的侧音；在泰恩赛德的情况下，只有清晰的支线；在 RP 和 GA 中，虽然有暗侧音和非暗侧音，但区别纯粹是同位音：区别在于单个音素的实现水平。因此，我们将这些类型的差异称为实现差异。

There are differences between accents which do not amount to a difference in the systems of phonological contrasts. Consider our discussion of dark and non-dark /l/ in chapter 7: we said there (7.6) that there is an allophonic rule in RP, to the effect that /l/ is realized as a velarized (‘dark’) lateral when it occurs in rhymes, but is realized as a non-velarized (‘non-dark’) lateral when it occurs in onsets. In many accents of the South of Scotland, in Australian English, in GA and in some accents of the North of England, /l/ is realized as a dark lateral in all positions, so that /lʌl/ (lull), for instance, is realized as [ɫʌɫ], rather than [lʌɫ]. In Tyneside English, on the other hand, /l/ is consistently realized as a clear lateral in all positions, so that /lʌl/ (lull)4 is realized as [ljʊlj]. There is no question of postulating, in these sorts of case, a difference in the underlying system of contrasts: all three accents have a contrast between /l/ and other consonant phonemes, such as /r/, but there is variation in the way that /l/ is realized. While there are languages in which the distinction between clear and dark laterals is contrastive, there can be no question of postulating such a phonological contrast in any of these accents of English: in the Southern Scottish and Australian cases, there are only dark laterals; in the Tyneside case, there are only clear laterals; and in RP and GA, while there are dark and non-dark laterals, the distinction is purely allophonic: the distinction lies at the level of realizations of a single phoneme. We will therefore refer to these sorts of difference as realizational differences.

涉及元音的实现差异非常普遍。例如，SSE 与大多数非苏格兰口音之间的差异之一是 SSE 中存在同位元音长度。这在大多数非苏格兰口音中是不存在的，但这种差异纯粹是现实问题。以 SSE 中 /i/ 和 /u/ 之间的对比为例（如beet 和 boot）：平行对比也存在于大多数非苏格兰口音中，包括 RP。在 RP 中，等效音素 /i:/ 和 /u:/ 在所有上下文中通常分别实现为 [i:] 和 [u:]，如在geese和goose中。然而，在 SSE 中，/i/ 被实现为短 [i]（如feet）或长 [i:]（如freeze), 并且 /u/ 被实现为短 [ʉ] (如foot ) 或长 [ʉ] (如lose ), 取决于音韵语境：长实现发生在词尾和浊音段 [ɹ]、[z]、[v]、[ð] 和 [ʒ] 之前；短期实现发生在其他地方。

Realizational differences involving vowels are very common. For instance, one of the differences between SSE and most non-Scottish accents is the presence of allophonic vowel length in SSE. This is absent in most non-Scottish accents, but this difference is a purely realizational matter. Take the contrast between /i/ and /u/ in SSE (as in beet, boot): a parallel contrast is also present in most non-Scottish accents, including RP. In RP, the equivalent phonemes, /i:/ and /u:/, are, in all contexts, typically realized as [i:] and [u:] respectively, as in geese and goose. In SSE, however, /i/ is realized as either short [i] (as in feet) or long [i:] (as in freeze), and /u/ is realized either as short [ʉ] (as in foot) or long [ʉ] (as in lose), depending on the phonological context: the long realization occurs word-finally and before the voiced segments [ɹ], [z], [v], [ð] and [ʒ]; the short realization occurs elsewhere.

随着时间的推移，实现差异可能会变成系统性差异。例如，在 RP 历史上的一个阶段，当它仍然是一个rhotic accent 时，/i:/ 音位在尾音位置 /ɹ/ 之前被实现为 [i:ə]，与 /i 几乎相同： / 目前在结尾位置的 /l/ 之前实现为 [i:ə]。在 RP 历史的那个阶段，feared/feed和beard/bead等对在两个方面有所不同：一个有 [ɹ]，而另一个没有，一个有 [i:]，而另一个有 [i :ə]。随着尾声位置 [ɹ] 发音的逐渐丢失，像这样的对变成最小对：[bi:əd] ( beard ) vs [bi:d] ( bead). 可以说，在那个阶段，出现了一个新的 /i:ə/ 音素（现代 /ɪə/ 的祖先）。这个过程被称为音位分裂：曾经是同位音的区别变成了音位。就口音之间的差异而言，我们想说的是，当 RP 和 SSE 都变音时，它们都有 /i:/ 音素，但在那个 RP 中存在实现差异，但不是SSE，有一个 /i:/ 音素的 [i:ə] 变位音。然而，现在存在系统性差异：RP 具有 /i:/ 与 /ɪə/ 对比，而 SSE 则缺乏。

Realizational differences can become systemic differences over time. For instance, at a stage in the history of RP when it was still a rhotic accent, the /i:/ phoneme was realized as [i:ə] before /ɹ/ in coda position, in much the same way as /i:/ is currently realized as [i:ə] before /l/ in coda position. At that stage in the history of RP, pairs such as feared/feed and beard/bead differed in two respects: one had [ɹ], while the other did not, and one had [i:], while the other had [i:ə]. With the gradual loss of the [ɹ] articulation in coda position, pairs like this became minimal pairs: [bi:əd] (beard) vs [bi:d] (bead). It is reasonable to say that, at that stage, a new /i:ə/ phoneme (the ancestor of present-day /ɪə/) emerged. This process is known as a phonemic split: a distinction which was once allophonic becomes phonemic. In terms of differences between accents, we want to say that, at a point when both RP and, say, SSE were rhotic, they both had the /i:/ phoneme, but there was a realizational difference, in that RP, but not SSE, had an [i:ə] allophone of the /i:/ phoneme. Now, however, there is a systemic difference: RP has an /i:/ vs /ɪə/ contrast which SSE lacks.

在 RP 和许多其他英语变体的历史中发生的音位分裂的另一个例子是上面提到的 FOOT/STRUT 分裂。STRUT 音素历史上是从 FOOT 音素演变而来的：在英语历史的某个阶段，FOOT 元音的所有实现都是圆音的。但是不圆润的实现开始出现，这些最终呈现出音位状态，导致出现了 STRUT 元音，在 RP 和许多其他英语变体中发音为 [ʌ]。英格兰北部的许多口音都没有经过 FOOT/STRUT 拆分，所以像strut这样的词发音为 [ʊ]。因此，这些英格兰北部口音和 RP 之间的区别是系统性的区别：RP 具有那些北方变体所没有的音位区别。因此，RP 中最小对的词对，例如put ([pʰʊt]) 和putt ([pʰʌt])，在北方变体中是同音字：put和putt都读 [pʰʊt]。

Another example of a phonemic split which occurred in the history of RP and many other varieties of English is the FOOT/STRUT split, mentioned above. The STRUT phoneme evolved historically from the FOOT phoneme: at one stage in the history of English, all realizations of the FOOT vowel were rounded. But unrounded realizations began to appear, and these eventually took on phonemic status, resulting in the emergence of the STRUT vowel, pronounced [ʌ] in RP and many other varieties of English. Many accents in the North of England failed to undergo the FOOT/STRUT split, so that words like strut are pronounced with an [ʊ]. The difference between these North of England accents and RP is therefore a systemic difference: RP possesses a phonemic distinction which is absent in those Northern varieties. As a result of this, pairs of words which are minimal pairs in RP, such as put ([pʰʊt]) and putt ([pʰʌt]), are homophones in the Northern varieties: both put and putt are pronounced [pʰʊt].

对于给定的差异，人们如何判断它是系统性的还是实现性的？让我们研究一个特殊的案例，即伦敦英语。我们所说的“伦敦英语”（以下简称 LE）并不是指所有伦敦本地人所说的口音；相反，我们指的是伦敦当地工人阶级的言论，尽管这种方式必然过于简单。⁵人们普遍注意到，LE 的说话者经常（但有所不同）说出lay、pay、say等带有 [aɪ] 双元音的词，我们知道，在 RP 中，这些词有一个 [eɪ] 双元音。这是系统性的还是实现性的差异？

How can one tell, for a given difference, whether it is systemic or realizational? Let us examine a particular case, that of London English. By ‘London English’ (henceforth, LE), we do not mean the accent spoken by all natives of London; rather, we refer, albeit in a necessarily oversimplified way, to the speech of working-class natives of London.5 It has been widely noted that speakers of LE often (but variably) utter words such as lay, pay, say with an [aɪ] diphthong, and we know that, in RP, these have an [eɪ] diphthong. Is this a systemic or a realizational difference?

答案是我们无法仅根据这些证据来判断。为了确定口音之间是否存在系统差异，我们必须考虑每种口音的对比系统。具体来说，我们必须要问：这两种口音中是否存在语音 [eɪ]/[aɪ] 的区别，如果有，它是否具有对比性？我们已经确定，在 RP 中，存在这样的区别，并且它是对比的（参见最小对bay vs buy，Tay vs tie，say vs sigh等）。然后我们必须确定这些对是否也是 LE 中的最小对：如果他们转向是 LE 中的同音词，那么在所有条件相同的情况下，我们可以合理地得出结论，这里存在系统性差异，就像我们根据来自最小对和同音词的证据得出的结论一样，RP 具有/ʊ/ 与 /ʌ/ 对比，这是许多北方英语口音（只有 /ʊ/）所缺乏的。

The answer is that we cannot tell on the basis of this evidence alone. In order to establish whether there is a systemic difference between the accents, we must consider the system of contrasts in each accent. Specifically, we must ask: is there a phonetic [eɪ]/[aɪ] distinction in either accent, and if so, is it contrastive? We have already established that, in RP, there is such a distinction, and that it is contrastive (cf. the minimal pairs bay vs buy, Tay vs tie, say vs sigh, etc.). What we must then establish is whether these pairs are also minimal pairs in LE: if they were to turn out to be homophones in LE, then we could, all things being equal, reasonably conclude that there is a systemic difference here, in just the same way that we concluded, on the basis of evidence from minimal pairs and homophones, that RP has an /ʊ/ vs /ʌ/ contrast, which many Northern English accents (with /ʊ/ alone) lack.

我们发现这些确实是 LE 中的最小对：而bay、Tay、say等在 LE 中有一个 [aɪ] 双元音，buy、tie、 sigh 等有一个 [ɔɪ] 双元音。在没有任何相反证据的情况下，我们可以得出结论，这里没有系统性差异：我们在 RP 中发现的对比在 LE 中得到保持。

What we find is that these are indeed minimal pairs in LE: while bay, Tay, say, etc. have an [aɪ] diphthong in LE, buy, tie, sigh, etc. have an [ɔɪ] diphthong. In the absence of any further evidence to the contrary, we may conclude that there is no systemic difference here: the contrast which we found in RP is maintained in LE.

但事情并没有就此结束，因为我们还知道 RP 在 /aɪ/ 和 /ɔɪ/ 之间有对比，例如buy, tie, sigh与boy, toy, soy。我们现在必须问的是，这种对比在 LE 中是否也存在，或者这些 RP 最小对是否在 LE 中是同音字。答案是对比在 LE 中持续存在：buy、tie、 sigh 等有 [ɔɪ]，boy、toy、soy等有 [oɪ]。

But the matter does not end there, since we also know that RP has a contrast between /aɪ/ and /ɔɪ/, as in buy, tie, sigh vs boy, toy, soy. What we must now ask is whether this contrast is also sustained in LE, or whether these RP minimal pairs are homophones in LE. The answer is that the contrast is sustained in LE: while buy, tie, sigh etc. have [ɔɪ], boy, toy, soy, etc. have [oɪ].

我们现在已经注意到 RP 和 LE 之间一组相关的非系统性（非对比性）纯实现差异，我们可以根据音位实现所在的发音和感知“空间”来设想这些差异。我们可以用元音空间图来描述这一点。考虑 RP 元音音素 /eɪ/、/aɪ/ 和 /ɔɪ/ 的语音实现。让我们想象一下，RP 的说话者通常对那些起点落在元音空间中以下类型的发音“区域”中的音素具有双元音实现：

We have now noted a related set of non-systemic (non-contrastive), purely realizational differences between RP and LE, which we might conceive of in terms of the articulatory and perceptual ‘space’ in which the realizations of a phoneme are located. We may depict this in terms of the vowel space diagram. Consider the phonetic realizations of the RP vowel phonemes /eɪ/, /aɪ/ and /ɔɪ/. Let us imagine that speakers of RP typically have diphthongal realizations of those phonemes whose starting points fall within the following sorts of articulatory ‘zones’ in the vowel space:

(2)

说话者对这些音素的理解可能会有所不同。但是，只要特定元音的发音不侵占任何其他元音的空间，bay仍可与buy和buy与boy区分开来。如果在口音的历史发展过程中，/eɪ/ 和 /aɪ/ 的发音变得如此接近以至于在感知上难以区分，以至于像bay和buy这样的对都发为 [baɪ]，那么对比度将丢失。这种在人类语言史上得到广泛证实的现象被称为音位合并: 一旦存在音位对比，它就会崩溃。^6个

The speaker might well vary in her or his realizations of these phonemes. However, so long as the articulation of a particular vowel does not encroach upon the space of any of the others, bay will still be distinguishable from buy, and buy from boy. If, in the course of the historical development of the accent, the articulations of /eɪ/ and /aɪ/ were to become so close as to be perceptually indistinguishable, such that pairs like bay vs buy were both uttered as [baɪ], then the contrast would be lost. That phenomenon, which is widely attested in the histories of human languages, is referred to as a phonemic merger: where once a phonemic contrast was present, it came to be collapsed.6

显然，在上面引用的 LE 案例中没有发生合并：虽然 /eɪ/ 的实现确实“转移”到 [aɪ]，但 [aɪ] 的实现又转移了，到 [ɔɪ]，从而保持对比度。类似地，当 /aɪ/ 的实现转变为 [ɔɪ] 时， /ɔɪ/ 的实现又转变为 [oɪ]，因此再次保持对比。⁷这种相当普遍的现象通常被称为元音移位. 澳大利亚英语中的平行移动很明显，其中 /i:/ 音位双元音化为 [ɪɪ]，因此在某种程度上侵占了 /eɪ/ 的空间。/eɪ/ 音素又转变为 [ɐɪ]，起点相当低，居中，不圆润，因此侵占了 /aɪ/ 的感知空间。反过来，/aɪ/ 变成了 [ɔɪ]，从而侵占了 /ɔɪ/ 的空间。但是，/ɔɪ/ 音素似乎没有采取“规避行动”，因此 [ɔɪ] 和 [ɔɪ] 是对比的。要记住的一点是，元音转换纯粹是实现的：它们不涉及语音对比数量的变化。

Clearly, no merger has occurred in the LE cases cited above: while the realization of /eɪ/ has indeed ‘shifted’ to [aɪ], the realization of [aɪ] has, in turn, shifted, to [ɔɪ], and the contrast is thus maintained. Similarly, while the realization of /aɪ/ has shifted to [ɔɪ], the realization of /ɔɪ/ has, in turn shifted, to [oɪ], and again the contrast has thus been maintained.7 This kind of phenomenon, which is fairly widespread, is often referred to as a vowel shift. A parallel shift is evident in Australian English, where the /i:/ phoneme has diphthongized to [ɪɪ], thus encroaching, to some extent, on the space of /eɪ/. The /eɪ/ phoneme has in turn shifted to [ɐɪ], with a fairly low, central, unrounded starting point, thus encroaching on the perceptual space of /aɪ/. In turn, /aɪ/ has shifted to [ɔɪ], thus encroaching on the space of /ɔɪ/. However, the /ɔɪ/ phoneme appears not to have taken ‘evasive action’, so [ɔɪ] and [ɔɪ] are contrastive. The point to be borne in mind is that vowel shifts are purely realizational: they do not involve a change in the number of phonological contrasts.

我们询问 /aɪ/ 作为 [ɔɪ] 的实现是否导致了 RP 中存在的对比的破坏。我们可能同样会问，在 LE 中将 /ɔɪ/ 音素实现为 [oɪ] 是否会破坏 RP 中的对比；答案是，尽管说 RP 的人可能会同时发出 [ɔɪ] 和 [oɪ]，但两者之间的语音差异虽然可以察觉，但从来都不是对比的：我们不能引用涉及这两者的最小对。所以不存在音位对比，如果将 /ɔɪ/ 说成 [oɪ] 会被破坏（RP 说话者可以自由地这样做，没有混淆对比的风险）。对于 LE 说话者，虽然 /eɪ/ 的实现侵占了 /aɪ/ 的“空间”，而后者的实现也侵占了 /ɔɪ/ 的空间，

We asked whether the realization of /aɪ/ as [ɔɪ] resulted in the destruction of a contrast which is present in RP. We might equally ask whether the realization of the /ɔɪ/ phoneme as [oɪ] in LE results in the destruction of a contrast found in RP; the answer is that, although RP speakers may well utter both [ɔɪ] and [oɪ], the phonetic difference between the two, while perceptible, is never contrastive: one cannot cite minimal pairs involving the two. So there is no phonemic contrast which would be collapsed by uttering /ɔɪ/ as [oɪ] (the RP speaker is free to do so, without risk of conflating a contrast). For the LE speaker, while the realization of /eɪ/ has encroached upon the ‘space’ of /aɪ/, and the realization of the latter has encroached upon the space of /ɔɪ/, the realization of the latter has not encroached upon the space of any other phoneme.

12.3 知觉和发音空间

12.3 Perceptual and Articulatory Space

世界语言中最简单的元音音位系统^{8是三元音 /i/、/u/、/a/ 系统，通常在元音空间内描述为：}

The simplest system of vowel phonemes8 found in the world’s languages is the three-vowel /i/, /u/, /a/ system, typically depicted within the vowel space as:

(3)

在这样的系统中，元音音素的实现可能会有很大差异；如果 /i/ 被实现为 [e] 型元音，就听者识别给定的元音音位而言，这将无关紧要，只要它相对靠前、不圆唇且比较高。/u/ 是否实现为 [o] 型元音也无关紧要，只要它相对较高、相对靠后且圆润即可。同样，/a/ 的变体是前面（如 [a]）、后面（如 [ɑ]）还是中央（如 [ɐ]）并不重要，只要它们相对较低且不圆润即可。也就是说，每个元音音素都有相当大的感知和发音空间。

The realizations of the vowel phonemes in a system like this can vary considerably; it will matter little in terms of the hearer’s identification of a given vowel phoneme if /i/ is realized as an [e]-type vowel, so long as it is relatively front, unrounded and relatively high. Nor will it matter if /u/ is realized as an [o]-type vowel, so long as it is relatively high, relatively back and rounded. Similarly, it will not matter whether realizations of /a/ are front (like [a]), back (like [ɑ]), or central (like [ɐ]), so long as they are relatively low and unrounded. That is, each vowel phoneme has quite a large perceptual and articulatory space.

世界语言中经常遇到的稍大一点的元音系统有中元音：/i/ /u/

A slightly larger vowel system, frequently encountered in the world’s languages, has mid vowels: /i/ /u/

(4)

在这样的系统中，元音空间有点拥挤：/i/ 的实现是否类似于 [e]，或者 /u/ 的实现是否类似于 [o]，这将很重要。但是 /e/ 的实现是否是 [ε]-like 并不重要，只要它不是 [a]-like。/o/ 的实现是否类似于 [ɔ] 也无关紧要。

In a system like this, the vowel space is a little more crowded: it will matter whether a realization of /i/ is [e]-like, or if a realization of /u/ is [o]-like. But it will not matter if a realization of /e/ is [ε]-like, so long as it is not [a]-like. Nor will it matter if a realization of /o/ is [ɔ]-like.

在稍大的系统中，高中元音和中低元音之间存在对比：

In a slightly larger system, there are contrasts between high-mid and low-mid vowels:

(5)

在这种系统中，每个元音音素的发音和感知空间甚至更少：例如，/e / 的实现是否类似于 [ε] 将很重要。

In this kind of system, there is even less articulatory and perceptual space for each vowel phoneme: it will matter if the realizations of /e / are [ε]-like, for instance.

许多英语口音具有相对较大的元音音素系统，例如通常包含多达三个或四个“a”型音素，如 RP 的前 /æ/、中 /ʌ/ 和后 /ɑ:/。这意味着实现的转变很容易导致一个音素侵占另一个音素的空间。这种表达和感知“空间”的概念与哪个问题有关在尝试确定给定的元音差异是否是系统的时，我们应该考虑的元音音素对。回答这个问题的一种方法是说应该考虑在发音和/或感知方面在某种意义上“相邻”的元音音素。这就是我们在考虑 [eɪ] 与 [aɪ] 区别时所做的：它们是相邻的，因为两个双元音的起点都在前面，非高位且不圆润。我们同样可以考虑 [ε] 与 [eɪ] 的区别，因为 [e] 是后者的起点，在发音方面接近 [ε]。事实上，人们发现 /ε/ 音素在 LE 中通常被实现为 [eɪ] 双元音（因此，好吧可以发音为 [weɪw])。然而，这不会导致 /ε/ 与 /eɪ/ 对比的合并，因为正如我们所见，/eɪ/ 被实现为 [aɪ]。

Many English accents have relatively large vowel phoneme systems, often containing, for instance, as many as three or four ‘a’-type phonemes, as in RP’s front /æ/, central /ʌ/ and back /ɑ:/. This means that shifts in realization can easily result in one phoneme encroaching upon the space of another. And this notion of articulatory and perceptual ‘space’ has a bearing on the question of which pairs of vowel phonemes we should consider when attempting to establish whether a given vowel difference is systemic or not. One way of answering this question is to say that one should consider vowel phonemes which are in some sense ‘adjacent’ in articulatory and/or perceptual terms. This is what we did when we considered the [eɪ] vs [aɪ] distinction: they are adjacent in that the starting points of both diphthongs are front, non-high and unrounded. We might equally have considered the [ε] vs [eɪ] distinction, since [e], the starting point for the latter, is close to [ε] in articulatory terms. Indeed, one finds that the /ε/ phoneme is often realized as an [eɪ] diphthong in LE (thus, well can be pronounced as [weɪw]). This, however, does not cause merger of the /ε/ vs /eɪ/ contrast since, as we have seen, /eɪ/ is realized as [aɪ].

在考虑相邻元音时遇到的主要困难是给定的元音可能与特定口音中发现的许多（可能是大多数）其他元音相邻。正如我们刚刚看到的，[eɪ] 与 [aɪ] 和 [ε] 相邻。它也与 [ɜ:] 相邻，因为 [eɪ] 的起点是前、中和不圆唇，而 [ɜ:] 是中央、中和不圆唇。碰巧的是，[ɜ:] 在 LE 中没有发生变化，但在其他口音中发生了变化。例如，在利物浦口音中，它前面是 [ε:]，因此work和bird, 在 RP 中发音为 [wɜ:k] 和 [bɜ:d]，在利物浦发音为 [wε:x] 和 [bε:d]。这意味着 [ɜ:] 的实现已经侵占了相邻元音音位 /ε/ 的发音和感知空间。然而，Liverpool 保留了 [ɜ:] 和 /ε/ 之间的音位区别，因为前者相对于后者来说很长，例如bird vs bed（[bε:d] vs [bεd]），因此可以与它。对于泰恩赛德郡的许多人来说，[ɜ:] 在某些词中被缩回并四舍五入，从而侵占了 [ɔ:] 的空间。因此，许多说话者将工作发音为 [wɔ:k]。这导致在一组特定的词中失去了 [ɜ:] 和 [ɔ:] 对比，所以像walk/work和鸟/无聊是同音词而不是最小对。

The main difficulty one encounters in considering adjacent vowel sounds is that a given vowel can be adjacent to many (perhaps most) of the other vowel sounds found in a particular accent. As we have just seen, [eɪ] is adjacent to both [aɪ] and [ε]. It is also adjacent to [ɜ:], since the starting point for [eɪ] is front, mid and unrounded, and [ɜ:] is central, mid and unrounded. As it happens, [ɜ:] has not shifted in LE, but it has in other accents. In the Liverpool accent, for instance, it has fronted to [ε:], so that work and bird, which are pronounced [wɜ:k] and [bɜ:d] in RP, are pronounced [wε:x] and [bε:d] in Liverpool. This means that the realization of [ɜ:] has encroached upon the articulatory and perceptual space of the adjacent vowel phoneme /ε/. However, the phonemic distinction between [ɜ:] and /ε/ is maintained in Liverpool, since the former is long with respect to the latter, as in bird vs bed ([bε:d] vs [bεd]) and thus distinguishable from it. For many Tyneside speakers, [ɜ:] has been retracted and rounded in some words, thus encroaching on the space of [ɔ:]. Thus work is pronounced [wɔ:k] by many speakers. This has resulted in a loss of the [ɜ:] vs [ɔ:] contrast in a particular set of words, so that pairs such as walk/work and bird/bored are homophones rather than minimal pairs.

元音发音似乎特别容易受到这种“左右移动”现象的影响，正因为如此，口音之间的大部分差异是元音发音的差异，而不是辅音。这可能是因为元音发音的性质，它们都有一个开放近似的限制。此外，元音越开，近似越开，舌头与口腔其他部分的接触越少。很明显，对于任何给定的语音片段，我们不太可能每次尝试都找到完全相同的发音。并非我们说出的每个 [d] 都会有完全相同的舌头部分关闭完全相同的牙槽嵴部分，持续时间完全相同，声带振动量完全相同，开始和结束时间完全相同. 所以变化是语音中固有的。

It appears that vowel articulations are especially susceptible to this ‘shifting around’ phenomenon and it is because of this that the majority of differences between accents are differences in the articulation of vowels, rather than of consonants. This is perhaps because of the nature of vowel articulations, which all have a stricture of open approximation. Furthermore, the more open the vowel, the more open the approximation, and the less contact there is between the tongue and the other parts of the oral cavity. It seems clear that, for any given phonetic segment, we are unlikely to hit on exactly the same articulation each and every time we attempt it. Not every [d] we utter will have exactly the same part of the tongue closing against exactly the same part of the alveolar ridge for exactly the same length of time with exactly the same amount of vocal cord vibration beginning and ending at exactly the same time. So variation is inherent in speech.

但它似乎在元音发音中比在塞音中更为固有，因为当一个人发出元音时比在一个人发出塞音时更难感觉到舌头在嘴里的位置。一个人可以感觉到嘴唇合拢以发出双唇停止的发音，一个人可以感觉到舌头当一个人发出牙槽塞音时，它靠在牙槽嵴上，这样说话者很容易辨别两者之间的发音差异。但是当一个人说出一个 [e:] 而不是一个 [ε] 时，就很难感觉到一个人的舌头在做什么。因此，一个人可以很容易地“过冲”或“下冲”，并以侵占另一个相邻元音音素空间的发音结束。简而言之，元音发音形成一个连续体；它们之间没有突然的、离散的划分。

But it appears to be even more inherent in vowel articulations than it is in, say, stops, since it is much harder to feel where one’s tongue is in one’s mouth when one produces a vowel than when one produces, say, a stop. One can feel the lips close together for the articulation of a bilabial stop, and one can feel the tongue against the alveolar ridge when one produces an alveolar stop, so that the articulatory difference between the two is easily discerned by the speaker. But when one produces, say an [e:] as opposed to an [ε], it is much harder to feel what one’s tongue is doing. One can therefore easily ‘overshoot’ or ‘undershoot’ and end up with articulations which encroach upon the space of another, adjacent, vowel phoneme. Vowel articulations, in short, form a continuum; there are no abrupt, discrete divisions between them.

与这个发音连续体平行的是一个感知连续体。回想一下，我们的“基数元音”仅仅是参考点，基于可用元音空间和舌头高度、前/后和圆度等可用参数的任意划分，所有这些都是程度问题，而不是一个绝对的问题。一旦 [e] 型声音开始降低，它在什么时候变成 [ε] 型声音？[æ]型的声音一旦开始升起，什么时候变成[ε]型的声音？答案是无法肯定地说。难怪，当一些说美式英语的人说bad时，说许多英式英语的人认为他们在说 bad：他们的 [æ] 型声音已经提升到被认为是 [ε] 型的声音。我们以感知颜色的方式感知元音。如果我们看到一个典型的绿色例子，我们可以毫不费力地将它识别为绿色。蓝色的经典示例也是如此。但是一旦我们看到一种介于两者之间的颜色，我们通常无法判断我们认为它是蓝绿色还是蓝绿色。在这种介于两者之间的情况下，一个人会判断一件衣服是绿色的，而另一个人会判断它是蓝色的。元音也是如此：我们发现很难在 [e] 和 [ε] 之间对元音进行分类。如果一个曾经接近原型 [ε] 的元音开始变得更接近 [e]，听者可能会将其归类为 [e] 的一个实例，因此可能反过来开始将其发音为 [e] ].

Parallel to this articulatory continuum, there is a perceptual one. Recall that our ‘cardinal vowels’ are merely reference points, based on an arbitrary carving up of the available vowel space and the available parameters of tongue height, front-ness/backness and roundedness, all of which are a matter of degree, rather than a matter of absolutes. Once an [e]-type sound begins to lower, at what point does it become an [ε]-type sound? Once an [æ]-type sound begins to raise, at what point does it become an [ε]-type sound? The answer is that it is impossible to say with any certainty. Little wonder, then, that when some American English speakers say bad, speakers of many accents of British English think that they are saying bed: their [æ]-type sound has raised to what is perceived as an [ε]-type sound. We perceive vowels the way we perceive colours. If we are presented with an example of a classic example of green, we have no difficulty in identifying it as green. The same is true for a classic example of blue. But once we are presented with a colour that is half-way between the two, we often cannot tell whether we think it is greenish-blue or bluish-green. In such in-between cases, one person will judge, say an item of clothing as green, while another will judge it to be blue. So it is with vowels: we find it hard to categorize a vowel half-way between [e] and [ε]. If a vowel that was once close to prototypical [ε] begins to be produced closer to [e], hearers may categorize it as an instance of an [e], and thus may, in turn, begin to articulate it as an [e].

最重要的一点是，我们对声音事件的感知在很大程度上取决于我们在母语中所拥有的语音对比的心理表征系统。同样重要的是，要存在音位区别，它必须基于人类可感知的语音区别。这些语音差异可能很小，但它们必须是可感知的。如果一种语言没有大量的元音音素库存，每个元音音素的实现就会在可用的发音和感知空间中“自由漫游”得多。

The most important point is that our perception of acoustic events is heavily dependent on the mentally represented system of phonological contrasts which we have in our native accent. It is also vital that, for a phonemic distinction to exist, it must be based on a phonetic distinction which is perceptible to human beings. Those phonetic differences can be minute, but they must be perceivable. If a language does not have a large vowel phoneme inventory, the realizations of each vowel phoneme are much more ‘free to roam’ in the available articulatory and perceptual space.

也有辅音连续体，但辅音之间的离散划分方式比元音之间多一点。例如，双唇发音与肺泡发音完全不同，因为舌头根本不涉及前者，而嘴唇不需要涉及后者。但即便如此，我们已经看到，尤其是在更像元音的辅音发音中，确实有可能从牙槽音逐渐过渡到唇音，正如我们在 /l/ 的 [w] 实现中看到的那样。一旦侧牙槽发音中出现次级软腭发音，它就可以变成初级发音，原来的牙槽狭窄可以完全消失，并出现新的双唇狭窄。

There are consonantal continua too, but there is a little more in the way of discrete divisions among consonants than among vowels. A bilabial articulation, for example, is radically distinct from an alveolar one, since the tongue is not implicated at all in the former, and the lips need not be involved in the latter. But even then, we have seen, particularly among consonantal articulations which are more vowel-like, that a gradual transition from alveolar to labial is indeed possible, as we have seen with [w] realizations of /l/. Once a secondary velar articulation is present in the pronunciation of an alveolar lateral, it can become the primary articulation, the original alveolar stricture can be lost altogether, and a new bilabial stricture can emerge.

12.4 音素词汇分布的差异

12.4 Differences in the Lexical Distribution of Phonemes

口音之间还有另一种变异，涉及单词的语音表示，但不是系统变异。例如，在英格兰北部的许多口音中，存在某种 /æ/ 与 /ɑ:/ 的区别，类似于 RP 中的区别。“long a”的实际元音质量通常与 RP 不同。与 RP 扬声器相比，许多西约克郡扬声器对“long a”的前置发音要多得多；区别通常在于短 [a] 和长 [a:] 之间，元音性质相同，两个元音仅在长度上有所区别。尽管如此，音位区别还是存在的，并且作为最小对的基础，如ant/aunt和Sam/psalm. 我们可以说西约克郡的 /a/ 与 /a:/ 的区别与 RP 的 /æ/ 与 /ɑ:/ 的区别平行或等同。

There is a further kind of variation between accents in which the phonological representation of words is concerned, but which is not systemic variation. For instance, in many accents of the North of England, there is some kind of /æ/ vs /ɑ:/ distinction, parallel to that found in RP. The actual vowel quality of the ‘long a’ often differs from that of RP. Many West Yorkshire speakers have a much more front articulation for ‘long a’ than RP speakers; very often the difference is between a short [a] and a long [a:], with the vowel qualities being the same, and the two vowels differentiated only with respect to length. Nonetheless, the phonemic distinction is there, and functions as the basis of minimal pairs such as ant/aunt and Sam/psalm. We can say that West Yorkshire’s /a/ vs /a:/ distinction is parallel to, or equivalent to, RP’s /æ/ vs /ɑ:/ distinction.

但是说这种北方英语口音（实际上也说 GA）的人通常会说出一个短元音，而说 RP 的人会说出一个长元音，即词汇集 BATH 的词。例如，对于许多北方人来说，bath、class和glass都以清擦音结尾。该组的其他成员具有鼻音加辅音序列，例如France和dance。关于这种差异，我们想说的不是说北方英语的人缺乏“long a”和“short a”的音位区别（就像苏格兰人所做的那样），而是这些特定单词的音韵形式包含短音素，而不是长音素。我们可以这样说明：

But speakers of such Northern English accents (and indeed of GA) often utter a short vowel in words which would be uttered with a long vowel by the RP speaker, words of the lexical set BATH. Examples, for many Northern speakers, are bath, class and glass, all ending with a voiceless fricative. Other members of the set have a nasal-plus-consonant sequence, such as France and dance. What we want to say about this sort of difference is not that the Northern English speaker lacks the ‘long a’ vs ‘short a’ phonemic distinction (in the way that Scottish speakers do) but that the phonological form of those particular words contains the short, rather than the long, phoneme. We can illustrate this as follows:

(6) 系统差异 vs 词汇分布差异

	音素	音韵形式
	'长/短'？	蚂蚁，阿姨，洗澡
RP 扬声器：	是的	/ænt/, /ɑ:nt/, /bɑ:θ/
北方人：	是的	/ant/, /a:nt/, /baθ/
苏格兰人：	不	/ɐnt/, /ɐnt/, /bɐθ/

这种差异很容易受到在其语音表示中具有一个音素而不是另一个音素的单词范围内不同口音之间的变化的影响。因此，与系统的和实现的差异相比，它们在本质上不那么普遍，而且更加特殊；然而，这种差异可能会导致不同方言使用者之间的相互理解出现重大问题。

Such differences are very susceptible to variation between different accents in the range of words which have one phoneme rather than the other in their phonological representations. They are therefore less general in nature, and more idiosyncratic, than systemic and realizational differences; however, such differences can cause major problems for mutual intelligibility between speakers of different dialects.

我们现在已经开始注意到口音和方言之间的一些差异，这种方式使我们能够比仅仅注意到不同的发音更深入地了解这些差异的本质，并表明理论考虑在我们的分析中发挥了多大作用。值得注意的是，英格兰北部的许多说话者将bus之类的词发音为 [bʊs]，而 LE 说话者将say之类的词发音为 [saɪ]，或者注意一些说话者在某些情况下会发出 [ɹ] 而不是与其他人相比，我们在这里所做的不仅仅是说一点：我们试图深入了解不同口音的说话者所拥有的语音知识的本质。

We have now begun to note some differences between accents and dialects, in a way which allows us more insight into the nature of those differences than merely noting different pronunciations, and which shows the extent to which theoretical considerations play a part in our analyses. While it is informative to note that many speakers in the North of England pronounce words like bus as [bʊs] and that LE speakers pronounce words like say as [saɪ], or to note that some speakers utter an [ɹ] in some contexts rather than others, what we have done here is to say a little more than just that: we have sought to gain some insight into the nature of the phonological knowledge possessed by speakers of different accents.

笔记

Notes

1这并不是否认社会对口音的态度会影响口音变化本身。显然，如果我们正在调查人们说话的方式，并且如果人们说话的方式受到社会对口音的态度的影响，那么我们就必须将这些因素视为总体情况的一部分。

1 This is not to deny that social attitudes to accents affect accent variation itself. Clearly, if we are investigating the way people speak, and if the way people speak is influenced by social attitudes to accents, then we are obliged to recognize those factors as part of the general picture.

2很多，但不是全部。一些北方口音在 /ʊ/ 和 /ʌ/ 之间有区别，其中 /ʌ/ 的实现是 [ʊ] 的不圆角版本，我们可以合理地将其转录为 [ɨ]。

2 Many, but not all. Some Northern accents have a distinction between /ʊ/ and /ʌ/, in which the realization of /ʌ/ is an unrounded version of [ʊ], which we might reasonably transcribe as [ɨ].

3另一个术语是“对比差异”。

3 An alternative term would be ‘contrastive difference’.

4许多讲泰恩赛德语的人缺乏 /ʊ/ 与 /ʌ/ 的区别。

4 Many Tyneside speakers lack the /ʊ/ vs /ʌ/ distinction.

5 “工人阶级”这个词含糊不清，而且非常难以定义。尽管如此，我们将假设它是有意义的，并且确实有助于识别社会阶层中真实的、复杂的差异（无论问题可能存在于“社会阶层”这个模糊的概念本身）。“伦敦”一词同样含糊不清。尽管如此，它也有一个有用的功能，因为它也使我们能够识别一个真正的，如果难以定义的地理实体，也许还有文化实体。

5 The term ‘working-class’ is vague, and enormously difficult to define. We will, nonetheless, assume that it is meaningful and does serve to identify what are real, if complex, differences in social class (whatever problems may reside in the vague notion ‘social class’ itself). The term ‘London’ is equally vague; nonetheless, it also serves a useful function, since it too allows us to identify a genuine, if hard-to-define, geographical, and perhaps cultural, entity.

6显然，如果一种实现与另一种实现之间存在差异，而语言学家没有注意到这一点，那么可能会被错误地认为已经合并的对比可能会重新出现。此外，对比可能会在某些语音环境中合并，但不会在其他语音环境中合并，从而使一些最小对保持完整，同时折叠其他对。参见下文 12.4 中关于词汇归属的内容。

6 Clearly, if there is variation between one realization and the other, and this is not noted by linguists, then a contrast that might have been wrongly taken to have been merged could re-emerge. Additionally, a contrast may merge in some phonetic environments, but not in others, leaving some minimal pairs intact while collapsing others. See 12.4 below on lexical incidence.

7我们在这里假设 LE 和 RP 有一个共同的来源，并且 LE 在历史上对这些音素的实现进行了创新，特别是在 RP 没有的方式上。这种假设在本案中恰好是有道理的，但不能假设所有的情况都是如此。在某些情况下，是 RP 进行了创新。/ʌ/ 与 /ʊ/ 对比就是这种情况，这是南方的创新：许多北方口音没有经历过这种创新。

7 We are assuming here that LE and RP have a common source, and that LE has innovated historically in its realizations of these phonemes in particular in a way which RP has not. This assumption happens to be justified in this case, but it must not be assumed that all cases are like this. There are cases in which it is RP which has innovated. Such is the case with the /ʌ/ vs /ʊ/ contrast, which is a Southern innovation: many Northern accents have not undergone that innovation.

8这里的“元音系统”是指“单元音系统”，忽略双元音。一般来说，单元音比双元音“更基本”，但双元音很常见，世界语言中最常见的是 [ai] 型和 [au] 型双元音。

8 By ‘vowel system’ here, we mean ‘system of monophthongs’, ignoring diphthongs. Generally speaking, monophthongs are ‘more basic’ than diphthongs, but diphthongs are common, and the most common in the world’s languages are the [ai]-type and [au]-type diphthongs.

练习

Exercises

gueqrecard_001

1在www.wiley.com/go/carrphonetics上收听曲目 12.1、12.2和12.3。对于每个声音文件，回答以下问题：

(a) 说话者是有皱纹的还是没有皱纹的？（参见第 7 章。）你有什么证据支持你的回答？

(b) 说话者在 LOT 之类的单词中是否有圆唇元音或不圆唇元音？

(d) 说话者是否表现出 Flapping 现象？（参见第 9 章。）其中一个是 RP 扬声器，另一个是 GA 扬声器，另一个是 SSE 扬声器。哪个是哪个？将你的回答与你对上述问题的回答联系起来。

2 如果说话者之间对 (1a–d) 中的现象存在差异，例如，对于每个差异，无论是系统差异还是实现差异，并解释原因。

13 一些英语口音的概述

An Outline of Some Accents of English

13.1 一些英国口音

13.1 Some British Accents

13.1.1 伦敦英语（位于www.wiley.com/go/carrphonetics的 轨道 13.4和练习 1）

13.1.1 London English (Track 13.4 at www.wiley.com/go/carrphonetics and exercise 1)

gueqrecard_001

13.1.1.1 定义伦敦英语口音

13.1.1.1 Defining the London English Accent

我们所说的“伦敦英语”是指在伦敦东区的自治市镇使用的白话、工人阶级的伦敦口音。这种口音的种类越来越多，范围从一端的伦敦腔到另一端在某些方面更接近 RP 的口音。

By ‘London English’ we mean the vernacular, working-class London accent spoken in the boroughs of the East End of London. There are more and less broad varieties of this accent, ranging from Cockney at one end of the spectrum to accents which are closer to RP in some respects at the other end.

13.1.1.2 伦敦英语元音

13.1.1.2 London English Vowels

让我们假设 LE 的元音系统与 RP 的元音系统完全平行，但有许多主要的实现差异，主要是“元音转换”性质，如第 12 章前面所述。

Let us postulate that the vowel system of LE is exactly parallel to that of RP, but with many major realizational differences, mostly of a ‘vowel shift’ nature, as discussed earlier in chapter 12.

/ʌ/、/æ/、/ε/、/eɪ/、/aɪ/ 和 /ɔɪ/ 的实现

Realizations of /ʌ/, /æ/, /ε/, /eɪ/, /aɪ/ and /ɔɪ/

在 LE 中，/ʌ/ 音素通常实现为短的 [a] 型声音。然后这会侵占 /æ/ 音素，这通常更接近 [ε]。这反过来又侵占了 /ε/ 音素，后者通常被认为是接近 [eɪ] 的双元音，而这种认识又侵占了 /eɪ/ 的感知空间。正如我们所见，/eɪ/、/aɪ/ 和 /ɔɪ/ 音素也参与了这种元音转换，我们可以将其描述如下：

In LE, the /ʌ/ phoneme is often realized as a short [a]-type sound. This then encroaches upon the /æ/ phoneme, which is often realized closer to [ε]. That in turn encroaches on the /ε/ phoneme, which in turn is often realized as a diphthong close to [eɪ], and that realization in turn encroaches on the perceptual space of /eɪ/. As we have seen, the /eɪ/, /aɪ/ and /ɔɪ/ phonemes also participate in this vowel shift, which we may depict as follows:

/aʊ/、/ʌ/、/æ/ 和 /ɑ:/ 的实现

Realizations of /aʊ/, /ʌ/, /æ/ and /ɑ:/

此外，在 LE 的伦敦版本中，/aʊ/ 音位通常实现为长 [a:]，如声音([sa:nd]) ¹和噘嘴([pa:t])。由于这个元音不同于 /ʌ/（如putt : [pat]）和 /æ/（如pat : [pεt]）的实现，因此保留了三个音素之间的区别。还要注意，由于 LE 中 /ɑ:/ 的实现没有改变，它与 /aʊ/ 的 [a:] 实现之间在元音质量上存在明显差异。

Additionally, in the Cockney version of LE, the /aʊ/ phoneme is often realized as a long [a:], as in sound ([sa:nd])1 and pout ([pa:t]). Since this vowel is distinct from the realizations of both /ʌ/ (as in putt: [pat]) and /æ/ (as in pat: [pεt]), the distinction between the three phonemes is preserved. Note too that, since the realization of /ɑ:/ in LE has not shifted, there is a clear difference in vowel quality between it and the [a:] realization of /aʊ/.

/ɔ:/ 和 /oʊ/ 的实现

Realizations of /ɔ:/ and /oʊ/

/ɔ:/ 在开音节中通常实现为 [ɔɐ]（如war 中：[wɔɐ]）和其他地方的 [oʊ]（如简称：[ʃoʊʔ]）。这意味着这个音素的一种实现侵占了/oʊ/音素的空间，导致音素重叠的可能性。反过来，该音素通常没有 [oʊ] 实现；相反，它倾向于在重复音节 /l/ 之前实现为 [ɒʊ]，在其他地方实现为 [ʌʊ]。

/ɔ:/ is typically realized as [ɔɐ] in open syllables (as in war: [wɔɐ]) and [oʊ] elsewhere (as in short: [ʃoʊʔ]). This means that one of the realizations of this phoneme encroaches upon the space of the /oʊ/ phoneme, leading to the possibility of phonemic overlapping. That phoneme in turn does not typically have an [oʊ] realization; rather, it tends to be realized as [ɒʊ] before a tautosyllabic /l/, and [ʌʊ] elsewhere.

音位重叠：/i:/ vs /ɪ/

Phonemic overlapping: /i:/ vs /ɪ/

/ɪ/ 音素通常在重复音节 /l/ 之前实现为 [ɪi]，如fill : [fɪiw]。

The /ɪ/ phoneme is often realized as [ɪi] before a tautosyllabic /l/, as in fill: [fɪiw].

类似地，/i:/ 音素通常在重复音节 /l/ 之前实现为 [ɪi]，如fill : [fɪiw]。因此，诸如此类的对在 LE 中是同音字，但在 RP 中是最小对。我们不认为 LE 经历了音位合并，因此这种差异本质上是系统性的，我们会说这是两个音位之间的音位重叠问题，因为在其他情况下，/i:/ 保留其 [ i:] 实现，而 /ɪ/ 保留其 [ɪ] 实现，如beat和bit：[bi:ʔ] 和 [biʔ]。请注意，当feel和fill末尾的 /l/被音节化为后面的开头时，如feeling和filling, /l/ 不再与前面的元音在同一个音节中，这是以其正常方式实现的，因此：[fi:lɪn] 和 [fɪ:lɪn]。

Similarly, the /i:/ phoneme is often realized as [ɪi] before a tautosyllabic /l/, as in fill: [fɪiw]. Thus pairs such as these are homophones in LE, but are minimal pairs in RP. Rather than concluding that LE has undergone a phonemic merger, and therefore that this difference is systemic in nature, we will say that it is a matter of phonemic overlapping between the two phonemes, since, in other contexts, /i:/ retains its [i:] realization, and /ɪ/ retains its [ɪ] realization, as in beat and bit: [bi:ʔ] and [biʔ]. Note that, when the /l/ at the end of feel and fill is syllabified into a following onset, as in feeling and filling, the /l/ is no longer in the same syllable as the preceding vowel, which is realized in its normal way, thus: [fi:lɪn] and [fɪ:lɪn].

降低单词结尾的 schwa

Lowering of word-final schwa

cinema和letter等词在 RP 中有词尾 schwa。在 LE 中，这个元音通常被降低到中心音 [ɐ]：字母通常发音为 [ˈlεʔɐ]。

Words such as cinema and letter have a word-final schwa in RP. In LE, this vowel is often lowered to a central [ɐ] sound: letter is often pronounced [ˈlεʔɐ].

13.1.1.3 伦敦英语辅音

13.1.1.3 London English Consonants

LE 在辅音音素的实现方面以及可以说在辅音音素系统方面都与 RP 不同。

LE differs from RP both in terms of the realization of consonant phonemes and, arguably, in terms of the consonant phoneme system.

清音停止音素

Voiceless stop phonemes

清音塞音 /p/、/t/、/k/ 通常在主重读元音之前带有沉重的送气音，至少在 /t/ 的情况下，通常带有摩擦音，一杯茶：[kʰaʔpətsʰɪi]。注意这些塞音的声门实现，如 [ʔ]，在比 RP 更广泛的上下文中。一个这样的语境是元音间（在元音之间），特别是当第一个元音被重读时，如matter : [ˈmεʔɐ]。由于第一个元音在这些词中是重读的，所以声门是脚内的。

The voiceless stop phonemes /p/, /t/, /k/ are often realized, before a primary stressed vowel, with heavy aspiration, and in the case of /t/ at least, often with affrication, cup of tea: [kʰaʔpətsʰɪi]. Note too glottal realizations of these stops, as [ʔ], in a wider range of contexts than in RP. One such context is intervocalically (between vowels), noticeably when the first vowel is stressed, as in matter: [ˈmεʔɐ]. Since the first vowel is stressed in such words, the glottaling is foot-internal.

/θ/ vs /f/ 和 /ð/ vs /v/ 在伦敦

/θ/ vs /f/ and /ð/ vs /v/ in Cockney

人们经常注意到，RP 最小对（例如thin/fin）对于许多说伦敦土话的人来说是同音词，都是 [fɪn]。这被 Wells 非正式地称为“TH-Fronting”（1982 年；参见推荐的进一步阅读）。/θ/ 是否可以说在 Cockney 中不存在在很大程度上取决于说 Cockney 的人在任何语境中是否说出 [θ]。如果发现 [f] 而不是 [θ]，总是出现在所有其他上下文中（例如在元音之间，如Cathy和词尾，如moth)，那么人们可以合理地得出结论，讲 Cockney 的人只是缺乏对比，并且 Cockney 和 RP 之间存在系统性差异。但是对于这种现象，许多说话人是可变的，所以我们不能断定他们缺少 /θ/ 音位。就 /v/ 与 /ð/ 的区别而言，很难找到涉及这两者的许多最小对（例如vs vat和live/lithe）。然而，人们已经注意到，像feather和with之类的词，在 RP 中分别具有元音间和词尾 /ð/，而在伦敦腔中通常用 [v] 发音。然而，词首 /ð/ 并不清楚，如the, that, there, their, this等，在伦敦腔中用 [v] 发音；如果这样的词是用 [ð] 说出来的，那么我们必须说这里存在纯粹的实现差异，Cockney /ð/ 在元音间和词尾被实现为 [v]。

It has often been noted that RP minimal pairs such as thin/fin are homophones for many Cockney speakers, both being [fɪn]. This is referred to informally as ‘TH-Fronting’ by Wells (1982; see Suggested Further Reading). Whether /θ/ may be said to be absent in Cockney depends very much on whether [θ] is uttered in any contexts by Cockney speakers. If one found that [f], rather than [θ], invariably turns up in all other contexts (e.g. between vowels, as in Cathy, and word-finally, as in moth), then one could reasonably conclude that Cockney speakers simply lack the contrast, and that there is a systemic difference here between Cockney and RP. But many speakers are variable with respect to this phenomenon, so we cannot conclude that they lack the /θ/ phoneme. As far as the /v/ vs /ð/ distinction is concerned, it is rather difficult to find many minimal pairs involving the two (that vs vat and live/lithe are examples). However, it has been noted that words such as feather and with, which have, respectively, intervocalic and word-final /ð/ in RP, are often uttered with [v] in Cockney. It is not clear, however, that word-initial /ð/, as in the, that, there, their, this, etc., is uttered with [v] in Cockney; if such words are uttered with [ð], then we must say that there is a purely realizational difference here, with Cockney /ð/ realized as [v] intervocalically and word-finally.

/l/ 发声

/l/ vocalization

/l/ 音素通常被认为是浊齿槽音近似音，但它可能会被次级发音覆盖，例如我们在 RP 音节的押韵位置发现的软化音，以及在重音中的首音和押韵位置，例如如苏格兰英语、GA 和澳大利亚英语。当次要发音成为主导时，牙槽发音可能会丢失，从而导致类似元音的发音。当在这个发音中添加圆唇时，在音节的结尾位置时，所产生的实现听起来非常像 [w] 声音。这可以在伦敦发音中看到，例如girl ([gεw])、Bethnal ([bεfnəw]) 和healthy([ˈεwfi])。这被称为“l 发声”，因为 /l/ 被实现为元音类型的声音。Cockney l-vocalization 似乎正在传播到英国的其他城镇。

The /l/ phoneme is often realized as a voiced alveolar approximant, but it may be overlaid by secondary articulations, such as the velarization we find in the rhyme position of the syllable in RP, and in both the onset and rhyme positions in accents such as Scottish English, GA and Australian English. When the secondary articulation becomes dominant, the alveolar articulation may be lost, resulting in a vowel-like articulation. When lip rounding is added to this articulation, the resulting realization sounds very much like a [w] sound, when in coda position in the syllable. This can be seen in Cockney pronunciations such as girl ([gεw]), Bethnal ([ˈbεfnəw]) and healthy ([ˈεwfi]). This is known as ‘l vocalization’ because the /l/ is realized as a vowel-type sound. Cockney l-vocalization seems to be spreading to other towns in the UK.

Cockney 中没有 /h/ 吗？

Absence of /h/ in Cockney?

人们普遍注意到，RP 最小对（例如hair/air）在 Cockney 中是同音字，也许在 LE 中通常是同音字。由于 Cockney 似乎甚至缺少单词内部 [h]，如behold所示，这看起来非常像系统差异。进一步的证据表明 /h/ 在伦敦腔中根本不存在，有两个来源。

It has been widely noted that RP minimal pairs such as hair/air are homophones in Cockney and perhaps generally in LE. Since Cockney appears to lack even word-internal [h], as in behold, this looks very much like a systemic difference. Further evidence that /h/ is simply absent in Cockney comes from two sources.

首先，我们知道[æn]（或[ən]）是不定冠词的读音形式，出现在英语中的元音词首词前，如ear、an oar等，而 [æ]（或 [ə]）是出现在以辅音开头的单词之前的形式，例如在船、房子等中。伦敦人在诸如house、hair等单词之前选择 [æn]（或 [ən]），等等，这可能被认为是暗示这些词在伦敦腔中是语音元音首字母。

Firstly, we know that [æn] (or [ən]) is the phonetic form of the indefinite article which occurs before vowel-initial words in English, as in an ear, an oar, etc., whereas [æ] (or [ə]) is the form which occurs before consonant-initial words, as in a boat, a house, etc. Cockneys select [æn] (or [ən]) before words such as house, hair, etc., which might be taken to suggest that such words are phonologically vowel-initial in Cockney.

其次，过度修正现象的证据很有说服力。当说话者过度正确时，他们会“纠正”不需要纠正的单词（尽量使它们接近被认为是在声望口音中发现的“正确”发音）。例如，许多通常会为后缀-ing说 [ɪn] 而不是 [ɪŋ] 的说话者可能仍然意识到 [ɪŋ] 而不是 [ɪn] 是“正确”的发音。这样的说话者经常会“纠正”单词，例如从 [kʰɪkɪn]踢到[kʰɪkɪŋ]，但也可能错误地将诸如badminton的单词“纠正”为 [bædmɪŋtən]。这种现象的意义在于，说话者将 [ɪn] 作为他或她的 - ing的音系形式词素，并将他或她的认识的“更正”过度概括为甚至不包含-ing后缀的情况。同样，缺少 /h/ 音素的法语使用者会过度纠正他们的英语，导致hair和air的发音如 [hεə] 。说法语的人的问题是，在没有 /h/ 音素的情况下，她或他不知道她或他的哪些没有 /h/ 的心理表征应该添加 /h/，哪些不应该添加。据观察，讲伦敦腔的人的行为方式与讲法语的人一样，将空气过度校正为 [hεə]，耳朵到 [hɪə] 等等。这强烈表明说伦敦伦敦人的人和说法语的人一样，根本就没有 /h/ 音素。然而，这种现象在伦敦英语中可能正在减少，这表明 LE 使用者中一直存在对 /h/ 音素的意识。

Secondly, evidence from the phenomenon of hyper-correction is rather telling. When speakers hyper-correct, they ‘correct’ words (try to make them approximate to what is considered the ‘proper’ pronunciation found in a prestige accent) which do not require correction. For instance, many speakers who would normally utter [ɪn] rather than [ɪŋ] for the suffix -ing may be nonetheless aware that [ɪŋ], rather than [ɪn], is the ‘correct’ pronunciation. Such speakers will often ‘correct’ words such as kicking from [kʰɪkɪn] to [kʰɪkɪŋ], but may also mistakenly ‘correct’ words such as badminton to [bædmɪŋtən]. The significance of this phenomenon is that the speaker has [ɪn] as his or her phonological form for the -ing morpheme, and overgeneralizes the ‘correction’ of his or her realizations to cases which do not even contain the -ing suffix. Similarly, speakers of French, who lack an /h/ phoneme, will hyper-correct their English, resulting in pronunciations such as [hεə] for both hair and air. The problem for the French speaker is that, in the absence of an /h/ phoneme, she or he is not to know which of her or his /h/-less mental representations should have an /h/ added and which not. Cockney speakers have been observed to behave in just the same way as French speakers, hyper-correcting air to [hεə], ear to [hɪə] and so on. This strongly suggests that Cockney speakers, like French speakers, simply do not have an /h/ phoneme. However, this phenomenon is perhaps in decline in London English, suggesting that awareness of the /h/ phoneme has always been present among LE speakers.

gueqrecard_002

13.1.2 泰恩赛德英语（曲目 13.5和练习 2）

13.1.2 Tyneside English (Track 13.5 and exercise 2)

13.1.2.1 定义泰恩赛德英语口音

13.1.2.1 Defining the Tyneside English Accent

“泰恩赛德英语口音”（也称为“乔迪口音”）是指泰恩河汇入北海之前最后几英里以北和南部城市地区的本地人所说的口音，主要包括泰恩河以北的纽卡斯尔和以南的盖茨黑德。

By ‘the Tyneside English accent’ (otherwise known as ‘the Geordie accent’), we mean the accent spoken by the natives of the urban areas to the north and south of the last few miles of the River Tyne before it meets the North Sea, including, principally, Newcastle upon Tyne to the north of the river and Gateshead to the south.

13.1.2.2 泰恩赛德英语元音

13.1.2.2 Tyneside English Vowels

/ʊ/ 与 /ʌ/

/ʊ/ vs /ʌ/

大多数泰恩赛德人都是北方人，没有 /ʊ/ 和 /ʌ/ 的区别：他们有前者的音素，但没有后者。²具有这种对比的口音经历了所谓的 FOOT/STRUT 分裂，这是一种历史变化，其中 /ʊ/ 音素发展为最终获得音素地位的不圆润的实现，因此像putt (/pʌt/) 和put (/pʊt/) 是最小对。在英格兰北部的口音中，例如 Geordie，这些通常是同音字，都发音为 [pʰʊt]。

Most Tyneside speakers are typically Northern in having no /ʊ/ vs /ʌ/ distinction: they have the former phoneme, but not the latter.2 Accents which have this contrast have undergone what is referred to as the FOOT/STRUT split, a historical change in which the /ʊ/ phoneme developed unrounded realizations which eventually gained phonemic status, so that pairs such as putt (/pʌt/) and put (/pʊt/) are minimal pairs. In North of England accents such as Geordie, these are typically homophones, both being pronounced [pʰʊt].

/e/ 和 /o/

/e/ and /o/

RP /eɪ/ 和 /oʊ/ 的泰恩赛德等价物是 /e/ 和 /o/。泰恩赛德郡的 /e/ 音素在许多泰恩赛德郡说话者的讲话中被实现为长单元音：[e:]。这种认识随着以 schwa 结尾的双元音认识而变化：[e:ə]。泰恩河畔的 /o/ 音素也可能被许多说话者理解为长单元音：[o:]。对于一些说话者来说，实现是一个长的单元音，它是 [ø:] 类型的前置 [o:]。

The Tyneside equivalents of RP /eɪ/ and /oʊ/ are /e/ and /o/. The Tyneside /e/ phoneme is realized, in the speech of many Tyneside speakers, as a long monophthong: [e:]. This realization varies with a diphthongal realization ending in schwa: [e:ə]. The Tyneside /o/ phoneme may also be realized as a long monophthong by many speakers: [o:]. For some speakers the realization is a long monophthong which is a fronted [o:], of the [ø:] sort.

/ɔ:/, /ɜ:/ 和非 Rhoticity

/ɔ:/, /ɜ:/ and non-rhoticity

泰恩赛德没有流变，这当然影响了元音系统的发展。某些泰恩赛德语使用者在某些词中缺少 /ɔ:/ 和 /ɜ:/ 之间的对比，因此work和walk等对是同音字：[wɔ:k]。具有“更广泛”泰恩赛德口音的说话者在walk和talk等词中保持 /ɔ:/ 和 /ɜ:/ 之间的区别，因此 /ɔ:/ 被实现为 [ɐ:] 和 /ɜ:/ 被实现为 [ɔ :]: [wɐ:k]（步行）与 [wɔ:k]（工作）。

Tyneside is non-rhotic and this has, of course, affected the development of the vowel system. Some Tyneside speakers lack a contrast between /ɔ:/ and /ɜ:/ in certain words, so that pairs such as work and walk are homophones: [wɔ:k]. Speakers with a ‘broader’ Tyneside accent maintain the distinction between /ɔ:/ and /ɜ:/ in words like walk and talk, so that /ɔ:/ is realized as [ɐ:] and /ɜ:/ realized as [ɔ:]: [wɐ:k] (walk) vs [wɔ:k] (work).

Schwa 和 non-rhoticity

Schwa and non-rhoticity

在居中双元音中，泰恩河畔的 /ɪə/ 音位通常实现为 [ɪʌ] 或 [ɪɐ]。同样的效果发生在 /ʊə/ 的实现中，如poor : [pʰʊɐ]。泰恩河畔的 /εə/ 音位通常被实现为一个长单元音：[ε:]。

Among the centring diphthongs, the Tyneside /ɪə/ phoneme is typically realized as [ɪʌ] or [ɪɐ]. The same kind of effect occurs in realizations of /ʊə/, as in poor: [pʰʊɐ]. The Tyneside /εə/ phoneme is typically realized as a long monophthong: [ε:].

与居中双元音 /ɪə/ 和 /ʊə/ 的泰恩赛德发音相关，在泰恩赛德，词尾位置的 schwa 音素 (/ə/) 可以相当 [ʌ]- 或 [ɐ]-like（即低中央不圆唇元音），但这取决于单词的历史。历史上 schwa 后跟 /r/ 的地方，它往往是 [ʌ] 或 [ɐ]，就像在dresser中一样。

Related to the Tyneside pronunciations of the centring diphthongs /ɪə/ and /ʊə/, in Tyneside, the schwa phoneme (/ə/) in word-final position can be rather [ʌ]- or [ɐ]-like (i.e. a low central unrounded vowel), but this depends on the history of the word. Where schwa was followed by an /r/ historically, it tends to be [ʌ] or [ɐ], as in dresser.

低不圆唇元音

Low unrounded vowels

泰恩河畔的 /æ/ 音素通常实现为 [ɐ]，但当其后跟浊词尾辅音时通常实现为长 [ɐ:]，如lad ([l ^j ɐ:d])，但不在lass ([l ^j æs])。

The Tyneside /æ/ phoneme is typically realized as [ɐ], but often realized as a long [ɐ:] when it is followed by a voiced word-final consonant, as in lad ([ljɐ:d]), but not in lass ([ljæs]).

/ɑ:/

尽管说泰恩赛德语的人和许多说北方语的人一样，在某些词中（例如bath ）经常使用 /æ/ 而不是 /ɑ:/ ，但这既没有反映出泰恩赛德语和 RP 之间的系统差异也没有实现差异。这是一种词汇分布差异：两个音素中的哪一个出现在给定的单词中。所讨论的词是属于 Wells (1982) 所指的词汇集 BATH 的词；其中包括诸如bath和class之类的词，其中元音后跟一个清擦音，以及诸如grant和France之类的词，其中元音后跟一个鼻音加辅音群。

Although Tyneside speakers, like many Northern speakers, often have /æ/ rather than /ɑ:/ in some words (e.g. bath), this reflects neither a systemic nor a realizational difference between Tyneside and RP. It is a lexical-distributional difference: a matter of which of the two phonemes appears in a given word. The words in question are words belonging to what Wells (1982) refers to as the lexical set BATH; these include words such as bath and class, where the vowel is followed by a voiceless fricative, and words such as grant and France, where the vowel is followed by a nasal-plus-consonant cluster.

/aɪ/ 实现

/aɪ/ realizations

尽管受制于变化（词汇和社会语言学），/aɪ/ 通常实现为 [aɪ] 或 [eɪ] 词尾和浊擦音之前，但更多地实现中央起点，如 [ʌɪ]，别处。效果类似于标准苏格兰英语中 /aɪ/ 上的苏格兰元音长度概括。

Although subject to variation (both lexical and sociolinguistic), /aɪ/ is often realized as [aɪ] or [eɪ] word-finally and before voiced fricatives, but realized with a more central starting point, as [ʌɪ], elsewhere. The effect is similar to that of the Scottish vowel length generalization on /aɪ/ in Standard Scottish English.

13.1.2.3 泰恩赛德英语辅音

13.1.2.3 Tyneside English Consonants

/h/ 和 /l/ 在泰恩赛德

/h/ and /l/ in Tyneside

尽管几乎所有英语口音都允许在功能类别（他、他等）的非重读词中不实现 /h/，但只有一些口音允许在功能类别的单词的重读音节中不实现 /h/词汇范畴。这些重音可以在英格兰的许多地方找到，但在泰恩赛德没有：/h/ 几乎总是在泰恩赛德的重读音节中实现。

Although almost every accent of English allows for non-realization of /h/ in unstressed words of a functional category (he, him, etc.), only some allow for non-realization of /h/ in the stressed syllables of words of a lexical category. These accents can be found in many parts of England, but not in Tyneside: /h/ is almost always realized in stressed syllables in Tyneside.

Tyneside /l/ 在所有位置都被实现为“清晰的 l”，转录为 [l ^j ]。在这种情况下，术语“清晰的 l”表示具有次级硬腭关节的牙槽侧向近似，其中舌头的前部与硬腭形成关节。

Tyneside /l/ is realized as a ‘clear l’ in all positions, transcribed [lj]. The term ‘clear l’, in this context, denotes an alveolar lateral approximant with a secondary palatal articulation, in which the front of the tongue forms an articulation with the hard palate.

/p/、/t/ 和 /k/ 的声门塞音和声门化

Glottal stop and glottalization of /p/, /t/ and /k/

清塞音 /p/、/t/ 和 /k/ 经常在元音之间发生声门化，特别是当第一个元音有主要或次要重音时，如clipper、fitter、hacker中。由此产生的实现可以转录为 [ʔp]、[ʔt]、[ʔk]。口腔中的发音与声门闭合同时发生。/t/ 的 [ʔt] 实现随 [ʔɾ] 的变化而变化：一个声门化的拍子。“声门增强”一词有时用于表示这种发音。响音可能介于元音和塞音之间，如grumpy, auntie, hankie中。

The voiceless stops /p/, /t/ and /k/ often undergo glottalization between vowels, particularly when the first vowel has primary or secondary stress, as in clipper, fitter, hacker. The resulting realization can be transcribed as [ʔp], [ʔt], [ʔk]. The articulations in the oral cavity occur simultaneously with the glottal closure. The [ʔt] realization of /t/ varies with [ʔɾ]: a glottalized tap. The term ‘glottal reinforcement’ is sometimes used to denote this kind of articulation. Sonorants may intervene between the vowels and the stop, as in grumpy, auntie, hankie.

在此处引用的单词中，我们可以将这种声门化的上下文定义为脚内（即在主要或次要重读元音和非重读元音之间），因此 /p/、/t/ 和 /k/ 在重读音节的开头，但如果它们出现在脚内侧，则声门化。

In the words cited here, we could define the context for this glottalization as foot-internal (i.e. between a primary or secondary stressed vowel and an unstressed vowel), so that /p/, /t/ and /k/ are aspirated at the beginning of a stressed syllable but glottalized if they occur foot-internally.

说泰恩赛德英语的人也表现出声门化，其中清音塞音，尤其是 /t/，被实现为声门塞音 [ʔ]。这与我们所说的声门化不同，因为这里的发音是声门塞音，口腔中没有额外的闭合。在许多英语口音（包括 RP）中都可以发现 Glottaling，但如果它出现在重读元音之后的元音间，它在感知上会更加突出，例如butter等词, 在泰恩赛德英语中可以用声门化 ([bʊtʔɐ]) 或声门化 ([bʊʔɐ]) 发音。在这个位置上使用喉音在英国是一种社会耻辱，但在不列颠群岛的乔迪语、伦敦英语、工人阶级苏格兰英语和许多其他英语口音中却非常普遍。有趣的是，声称声门处理“懒惰”、“不清楚”或“邋遢”的 RP 说话者并没有意识到他们自己在大范围内发出声门塞音，但通常是在他们在感知上不太突出的语音环境中，如词尾声门在非重读虚词中停止，后跟词首辅音，在像I didn't know that she was here 这样的句子中，最后的 /t/很可能会作为声门塞音发出，即使是 RP 扬声器也是如此。具有讽刺意味的是，正是因为声间的、脚内的声门停止在诸如黄油之类的单词发音非常清晰（即清晰可闻），以至于它是那些说它不清楚的人抱怨的对象。

Speakers of Tyneside English also exhibit glottaling, in which voiceless stops, especially /t/, are realized as the glottal stop [ʔ]. This is distinct from what we have called glottalization, since the articulation here is a glottal stop with no additional closure in the oral cavity. Glottaling is found in many accents of English, including RP, but it is perceptually more salient if it occurs intervocalically after a stressed vowel, as in words such as butter, which in Tyneside English can be pronounced with glottalization ([ˈbʊtʔɐ]) or with glottaling ([ˈbʊʔɐ]). Glottaling in this position is socially stigmatized in Britain, but it is perfectly common in Geordie, London English, working-class Scottish English and many other accents of English in the British Isles. RP speakers who claim that glottaling is ‘lazy’, ‘unclear’ or ‘slovenly’ are, amusingly, unaware that they themselves utter glottal stops on a vast scale, but often in phonological contexts in which they are less salient perceptually, as in the case of word-final glottal stops in unstressed function words followed by a word-initial consonant, in sentences like I didn’t know that she was here, where the final /t/ of that is very likely to be uttered as a glottal stop, even by RP speakers. Ironically, it is precisely because the intervocalic, foot-internal glottal stop in pronunciations of words such as butter is so clear (i.e. clearly audible) that it is the object of complaints by those who say it is unclear.

/t/ 的 'r' 实现

The ‘r’ realizations of /t/

与许多英格兰北部口音一样，泰恩赛德有一个 /t/ 的实现，它是 [ɹ] 型或 [ɾ] 型发音。这种现象被非正式地称为“T-to-R”。目前尚不完全清楚是否存在发生这种情况的可陈述的语音语境，或者是否仅存在通常发生这种情况的词库或什至短语。这种现象可能在社会语言学上是可变的。这种认识让人想起 GA 中的 Flapping，因为它似乎发生在语音间，但 Tyneside 现象在词汇上更为零星。典型的情况似乎涉及一个词尾 /t/，前面是一个短元音，后面是一个以元音开头的词，如got a light, get off, put it down，但他确实是，闭嘴。但是，它可以出现在单词内部，如更好，在长元音之后，就像我认为的那样。'r' 的实现似乎也随着声门和声门化的实现而变化，因此给定的发音，比如说，下车可以有 [t]、[ʔ]、[ʔt]、[ʔɾ]、[ɾ] 或 [ɹ] ] 作为get中 /t/ 的实现。

Tyneside, like many North of England accents, has a realization of /t/ which is either an [ɹ]-type or an [ɾ]-type articulation. The phenomenon is known informally as ‘T-to-R’. It is not entirely clear whether there is a stateable phonological context in which this occurs, or whether there is simply a stock of words, or even phrases, in which it typically occurs. The phenomenon is probably sociolinguistically variable. The realization is reminiscent of Flapping in GA, in that it seems to occur inter-vocalically, but the Tyneside phenomenon is lexically much more sporadic. Typical cases seem to involve a word-final /t/ preceded by a short vowel and followed by a word beginning with a vowel, as in got a light, get off, put it down, but he does, shut up. It can, however, occur word-internally, as in better, and after long vowels, as in I thought he did. The ‘r’ realization also seems to vary with the glottal and glottalized realizations, so that a given pronunciation of, say, get off can have [t], [ʔ], [ʔt], [ʔɾ] [ɾ] or [ɹ] as the realization of the /t/ in get.

gueqrecard_002

13.1.3 标准苏格兰英语（曲目 13.6和练习 3）

13.1.3 Standard Scottish English (Track 13.6 and exercise 3)

13.1.3.1 定义标准苏格兰英语口音

13.1.3.1 Defining the Standard Scottish English Accent

标准苏格兰英语 (SSE) 是许多苏格兰人在说标准英语方言时所说的标准口音。这是受过大学教育的中产阶级苏格兰人的特征。它不同于所谓的苏格兰人，后者源自古英语的诺森伯兰方言。苏格兰人与苏格兰的工人阶级说话者有关；在欧文·威尔士 (Irvine Welsh)的小说《猜火车》和同名电影中都可以找到它的形象。许多说苏格兰语的人可以说标准英语而没有任何苏格兰语痕迹，但他们也可以将苏格兰语单词，如bairn（孩子）、lug（耳朵）和kirk（教堂）混入他们的标准英语中。

Standard Scottish English (SSE) is the standard accent which many Scots speak when speaking the Standard English dialect. It is characteristic of university-educated, middle-class Scottish speakers. It is distinct from what is referred to as Scots, which is derived from the Northumbrian dialect of Old English. Scots is associated with working-class speakers in Scotland; representations of it can be found in the novel Trainspotting, written by Irvine Welsh, and in the film of that name. Many Scottish speakers can speak Standard English without any trace of Scots, but they can also mix Scots words, such as bairn (child), lug (ear) and kirk (church), into their Standard English.

13.1.3.2 SSE 元音

13.1.3.2 SSE Vowels

苏格兰元音长度规则

The Scottish Vowel Length Rule

SSE 的一个主要特征是几乎没有音位元音长度的证据：/u:/ 与 /ʊ/、/ɑ:/ 与 /æ/ 和 /ɔ:/ 与 /ɒ/ 等对不构成元音音位系统。但是有相当多的证据表明变位元音长度。正如我们所见（第 142 页），一些元音音素在浊音连续体之前或末尾有长的同位异音，产生长/短同位异音差异，例如 [ɫif]/[ɫi:v] ( leaf/leave ) 和 [ hʉf]/[mʉ:v]（蹄/移动）。这种现象被称为苏格兰元音长度规则 (SVLR)，这是 SSE 和 RP 之间的主要实现差异。哪一组元音经过 SVLR 是一个有争议的问题；但是 /i/、/u/ 和 /ai/ 似乎几乎所有 SSE 都经历过它扬声器。/ai/ 的两个实现是 [ɐe]，如eye和wise，以及 [ʌi]，如right和ice。虽然 /ai/ 在与 /i/ 和 /u/ 的长短同位素完全相同的上下文中具有这两个异位音位，但“短”([ʌi]) 和“长”([ɐe]) 异位音位之间的区别似乎是元音质量问题，而不是数量问题。

A major characteristic of SSE is that there is little evidence of phonemic vowel length: pairs such as /u:/ vs /ʊ/, /ɑ:/ vs /æ/ and /ɔ:/ vs /ɒ/ do not form part of the vowel phoneme system. But there is considerable evidence for allophonic vowel length. As we have seen (p. 142), some of the vowel phonemes have long allophones morpheme-finally or before voiced continuants, yielding long/short allophonic differences such as [ɫif]/[ɫi:v] (leaf/leave) and [hʉf]/[mʉ:v] (hoof/move). This phenomenon is known as the Scottish Vowel Length Rule (SVLR), a major realizational difference between SSE and RP. Which set of vowels undergoes the SVLR is a matter of debate; but /i/, /u/ and /ai/ seem to undergo it for nearly all SSE speakers. The two realizations of /ai/ are [ɐe], as in eye and wise, and [ʌi], as in right and ice. While /ai/ has these two allophones in exactly the same contexts as the long and short allophones of /i/ and /u/, the difference between the ‘short’ ([ʌi]) and ‘long’ ([ɐe]) allophones seems to be a matter of vowel quality, rather than quantity.

/ʊ/ 与 /u:/ 和 /ʊ/ 与 /ʌ/ 的区别

The /ʊ/ vs /u:/ and /ʊ/ vs /ʌ/ distinctions

RP 和 SSE 之间的主要系统差异之一是 SSE 没有pool/pull（或 /u:/ vs /ʊ/）类型的区别。由于 SSE 没有音位长元音，因此没有 /u:/ 或 /ʊ/；相反，只有一个音素：/u/。这在 SVLR 上下文中实现为长 [ʉ:]，在其他地方实现为短 [ʉ]。在除 SSE 以外的苏格兰口音中，此元音的实现甚至可能比中央高音 [ʉ] 更靠前，有时接近法语类型的 [y] 音（前圆角高音元音）。

One of the major systemic differences between RP and SSE is that SSE does not have the pool/pull (or /u:/ vs /ʊ/) type of distinction. Since SSE does not have phonemically long vowels, there is no /u:/ or /ʊ/; instead, there is a single phoneme: /u/. This is realized as long [ʉ:] in the SVLR contexts, and realized as short [ʉ] elsewhere. In Scottish accents other than SSE, the realization of this vowel can be even more fronted than the high central [ʉ], sometimes approaching a French-type [y] sound (a high front rounded vowel).

与英格兰北部的许多口音不同，SSE确实具有put/putt（RP 中的 /ʊ/ 与 /ʌ/）类型的区别。因此，put、putt、pool和pull等词被实现为 [pʰʉt]（put）、[pʰʌt]（putt）和 [pʰʉɫ]（pull/pool），像poor这样的词被实现为 [pʰʉ:ɹ]，长元音，由 SVLR 触发。

SSE, unlike many North of England accents, does have the put/putt (/ʊ/ vs /ʌ/ in RP) type of distinction. The words put, putt, pool and pull are therefore realized as [pʰʉt] (put), [pʰʌt] (putt) and [pʰʉɫ] (pull/pool), with words like poor realized as [pʰʉ:ɹ], with a long vowel, triggered by the SVLR.

没有 /ɔ:/ 与 /ɒ/ 对比

Absence of the /ɔ:/ vs /ɒ/ contrast

RP 和 SSE 之间的另一个主要系统差异在于 /ɔ:/ 与 /ɒ/ 对比在 SSE 中缺失，取而代之的是单个音素：/ɔ/。这意味着，对于许多 SSE 说话者来说，cot/caught和not/nought等 RP 最小对是同音词。由于许多说 SSE 的人从英式英语中借用 /ɔ:/ 与 /ɒ/ 对比，因此情况变得复杂，因此 /ɔ:/ 在诸如naught和caught之类的词中，但 /ɔ/ 在诸如不和婴儿床。

Another major systemic difference between RP and SSE lies in the fact that the /ɔ:/ vs /ɒ/ contrast is missing in SSE, which has instead a single phoneme: /ɔ/. This means that, for many SSE speakers, RP minimal pairs such as cot/caught and not/nought are homophones. The picture is complicated by the fact that many SSE speakers have borrowed the /ɔ:/ vs /ɒ/ contrast from Anglo-English and thus have /ɔ:/ in words such as nought and caught, but /ɔ/ in words such as not and cot.

RP 在 Wells (1982) 称之为词汇集 THOUGHT、NORTH 和 FORCE 中有 /ɔ:/ 音素，但在词汇集 LOT 的单词中有音素 /ɒ/。我们已经看到，说话没有受到 RP 影响的 SSE speakers 没有这个 /ɔ:/ vs /ɒ/ 对比，所以caught（属于集合 THOUGHT）和cot（属于集合 LOT）是同音字。RP 和 SSE 在这些词汇集方面的另一个区别是 SSE 在 FORCE 词集的词中有 /o/ 元音，而在 NORTH 词集的词中有 /ɔ/ 元音。因此，在 SSE 中，有一些对，例如horse（属于集合 NORTH）和hoarse（属于集合 FORCE），它们是最小对：horse发音为 [hɔɹs]，而嘶哑则发音为 [hoɹ; 在 RP 中，这些是同音词，都发音为 [hɔ:s]。

RP has the /ɔ:/ phoneme in what Wells (1982) refers to as the lexical sets THOUGHT, NORTH and FORCE, but the phoneme /ɒ/ in words of the lexical set LOT. We have seen that SSE speakers whose speech has not been influenced by RP do not have this /ɔ:/ vs /ɒ/ contrast, so that caught (which belongs to the set THOUGHT) and cot (which belongs to the set LOT) are homophones. Another difference between RP and SSE with respect to these lexical sets is that SSE has the /o/ vowel in words of the set FORCE, but the /ɔ/ vowel in words of the set NORTH. As a result, in SSE there are pairs such as horse (which belongs to the set NORTH) and hoarse (which belongs to the set FORCE) which are minimal pairs: horse is pronounced [hɔɹs], while hoarse is pronounced [hoɹ; in RP, these are homophones, both being pronounced [hɔ:s].

没有 /æ/ 和 /ɑ:/ 区别

Absence of the /æ/ vs /ɑ:/ distinction

RP 和 SSE 之间另一个显着的系统差异在于 SSE 没有 /æ/ 与 /ɑ:/ 区别。相反，它有一个单一的 'a' 音素，实现为低不圆唇中央元音 [ɐ]，因此ant和aunt等对在 SSE 中是同音字：[ɐnt]。当然，这些在 RP 中是不同的：ant在 RP中发音为 [ænt]，而aunt发音为 [ɒ:nt]。一些受过教育的 SSE 演讲者有从 RP 中借用了这种对比，但它们是否产生这种对比通常是可变的。

Another striking systemic difference between RP and SSE lies in the fact that SSE does not have the /æ/ vs /ɑ:/ distinction. Instead, it has a single ‘a’ phoneme, realized as a low unrounded central vowel, [ɐ], so that pairs such as ant and aunt are homophones in SSE: [ɐnt]. These are, of course, distinct in RP: ant is pronounced [ænt] in RP, while aunt is pronounced [ɒ:nt]. Some educated SSE speakers have borrowed this contrast from RP, but are often variable in whether they produce it or not.

/eɪ/ 和 /oʊ/ 型音素

The /eɪ/ and /oʊ/-type phonemes

RP /eɪ/ 音素的 SSE 等价物是 /e/，有人说在 SVLR 上下文中实现为长单音 [e:]，在其他地方实现为短 [e]，如 [bet]（诱饵）和[be:ɹ]（光秃秃的）。

The SSE equivalent of the RP /eɪ/ phoneme is /e/, said by some to be realized as long monophthongal [e:] in the SVLR contexts, and as short [e] elsewhere, as in [bet] (bait) and [be:ɹ] (bare).

RP /oʊ/ 音素的 SSE 等价物是 /o/，有人说在 SVLR 语境中实现为长单音 [o:]，而在其他地方则称为短 [o]，如 [bot] ( boat ) 和 [ bo:ɹ]（野猪/钻孔）。

The SSE equivalent of the RP /oʊ/ phoneme is /o/, said by some to be realized as long monophthongal [o:] in the SVLR contexts, and short [o] elsewhere, as in [bot] (boat) and [bo:ɹ] (boar/bore).

双元音

The diphthongs

正如我们所见，SSE /ai/ 双元音经过 SVLR，在元音长度上下文中实现为 [ɐeɪ]，在其他地方实现为 [ʌi]。SSE /au/ 双元音实现为 [ʌʉ]，如口语：[mʌʉθ]。

As we have seen, the SSE /ai/ diphthong undergoes the SVLR, being realized as [ɐeɪ] in the vowel length contexts and [ʌi] elsewhere. The SSE /au/ diphthong is realized as [ʌʉ], as in mouth: [mʌʉθ].

/ə/、/ʌ/ 和 /e/

/ə/, /ʌ/ and /e/

许多在 RP 中具有词尾 schwa 且历史上不以 /ɹ/ 结尾的词，例如cinema、comma、America，在 SSE 中以 [ʌ] 型元音发音。在 RP 词尾 schwa 历史上后跟 /ɹ/ 的地方，SSE 保留 /ɹ/ 并且有一个 schwa 或 /ɪ/，如better和seller。

Many words which, in RP, have word-final schwa and did not historically end in an /ɹ/, such as cinema, comma, America, are uttered with an [ʌ]-type vowel in SSE. Where RP word-final schwa was historically followed by an /ɹ/, SSE retains the /ɹ/ and has either a schwa or an /ɪ/, as in better and seller.

在 RP 和许多其他口音中，在very、happy、lucky等词的末尾发现的短 [i] 元音通常是 SSE 中的 [e]。这个元音在 RP 中曾经是 [ɪ]，但经历了 Wells (1982) 所说的 happY Tensing，其中 [ɪ] 被提升或紧张成为 [i]。许多英语变体表现出快乐的紧张，但不是 SSE。

The short [i] vowel found, in RP and many other accents, at the ends of words such as very, happy, lucky is usually an [e] in SSE. This vowel used to be [ɪ] in RP, but has undergone what Wells (1982) refers to as happY Tensing, in which the [ɪ] has been raised, or tensed, to become [i]. Many varieties of English exhibit happY Tensing, but not SSE.

13.1.3.3 SSE 辅音

13.1.3.3 SSE Consonants

粗糙度

Rhoticity

SSE 很活跃；SSE /ɹ/ 音素出现在所有音节位置，包括尾音位置，因此像far和farm这样的词总是发音为“r”。音素通常实现为 [ɹ]，有时实现为 [ɾ]，很少实现为 [r]。一些说话者在 [ɹ] 和 [ɾ] 实现之间表现出异音区别，在分支首字母中实现轻拍，如bring、trip和creep等词。请注意，当我们谈到rhotic accent时，我们并不是指说话者发出的那种“r”音：我们指的是一种口音，其中“r”，无论它的语音形式是什么，都在尾声中实现位置。

SSE is rhotic; the SSE /ɹ/ phoneme is realized in all syllabic positions, including the coda position, so that words like far and farm are always pronounced with an ‘r’. The phoneme is typically realized as [ɹ], sometimes as [ɾ], and very rarely as [r]. Some speakers exhibit an allophonic distinction between the [ɹ] and [ɾ] realizations, with the tap being realized in branching onsets, in words such as bring, trip and creep. Note that, when we speak of a rhotic accent, we are not referring to the kind of ‘r’ sound a speaker has: we are denoting an accent in which the ‘r’, whatever its phonetic form might be, is realized in coda position.

/ʍ/ 与 /w/ 的区别

The /ʍ/ vs /w/ distinction

/ʍ/ 与 /w/ 的区别，如witch/which、weals/wheels和watt/what，正如拼写所暗示的那样，这种区别在 RP 中已基本消失，但在 SSE 和一些美国口音中仍然存在. /ʍ/ 音位被实现为清双唇擦音，具有次级软腭发音。

The /ʍ/ vs /w/ distinction, as in witch/which, weals/wheels and watt/what, as the spelling suggests, is a distinction which has been largely lost in RP, but is still present in SSE and some American accents. The /ʍ/ phoneme is realized as a voiceless bilabial fricative, with a secondary velar articulation.

/h/ 与 /x/ 与 /k/ 的区别

The /h/ vs /x/ vs /k/ distinctions

RP 和 SSE 之间的一个主要系统差异在于 SSE 保留了 /k/ 和 /x/ 之间的音位区别。/x/ 音位在押韵中表现为清软颚擦音 ([x])，位于低元音和后元音之后，如在loch（湖）中。它出现在许多苏格兰地名中，例如Auchertmuchty和Lochalsh。前高音位元音 ([ç]) 可能出现在高前元音之后，但这往往仅限于某些苏格兰语的单词，这些单词已被纳入 SSE 说话者的语音中；一个例子是dreich，发音为 [dɹiç] 或 [dɾiç]，这个词用来指寒冷、灰色和潮湿的天气。与 /ʍ/ 一样，/x/ 音位在 RP 中丢失；说 RP 的人经常说 [k] 而不是像loch这样的词. 与讲英语某些口音的人不同，讲 SSE 的人不会省略重读元音前的 /h/ 音素。

A major systemic difference between RP and SSE lies in the fact that SSE has retained the phonemic distinction between /k/ and /x/. The /x/ phoneme is realized as a voiceless velar fricative ([x]) in rhymes, after low vowels and back vowels, as in loch (lake). It occurs in many Scottish place-names, such as Auchertmuchty and Lochalsh. A fronted allophone ([ç]) may occur after high front vowels, but this tends to be restricted to certain words from Scots which have been incorporated into the speech of SSE speakers; an example is dreich, pronounced [dɹiç] or [dɾiç], a word used to refer to cold, grey and wet weather. As with /ʍ/, the /x/ phoneme has been lost in RP; RP speakers often utter [k] instead in words like loch. Unlike speakers of some accents in England, SSE speakers do not elide the /h/ phoneme before stressed vowels.

'黑暗'

‘Dark l’

SSE /l/ 音素在所有上下文中都被实现为“dark l”，即 [ɫ]。回想一下，'dark l' 是一个非正式术语，用于描述具有软腭化的次要关节的牙槽骨外侧近似。还要回想一下，“dark l”的实现只出现在 RP 中音节的韵脚中，因此像lull这样的单词中的两个 /l/ 音素的实现在 RP 中是不同的：[lʌɫ]。在 SSE 中，这个词读作 [ɫʌɫ]。

The SSE /l/ phoneme is realized as a ‘dark l’, i.e. [ɫ], in all contexts. Recall that ‘dark l’ is an informal term for an alveolar lateral approximant with a secondary articulation of velarization. Recall too that the ‘dark l’ realization only occurs in the rhyme of the syllable in RP, so that the realizations of the two /l/ phonemes in a word such as lull are distinct in RP: [lʌɫ]. In SSE, this word is pronounced [ɫʌɫ].

13.2 两种美国口音

13.2 Two American Accents

13.2.1 纽约市英语

13.2.1 New York City English

13.2.1.1 定义纽约市英语口音

13.2.1.1 Defining the New York City English Accent

纽约市的英语口音在地理上的定义相当明确，主要局限于纽约市的行政区（以下简称：纽约）。然而，纽约内部存在相当大的由社会决定的差异，而这种差异一直是大量社会语言学研究的主题。纽约口音在美国得到广泛认可，并且与许多城市口音一样，主要引起负面反应。在纽约进行的社会语言学研究中解决的问题之一是 rhoticity。很明显，口音已经实现了从变音到非变音的历史转变，因为它具有（大部分）以 schwa 结尾的居中双元音集，这是非变音口音的特征。然而，在振动方面存在相当大的社会语言学差异，似乎结尾位置的 [ɹ] 正在卷土重来。回想一下，美国的标准口音 General American 是rhoticity，并且rhoticity 被认为比非rhoticity 更有声望。

The New York City English accent is fairly sharply defined in geographical terms, being largely confined to the boroughs of New York City (henceforth: New York). There is, however, considerable socially determined variation within New York, and this variation has been the subject of a good deal of sociolinguistic study. The New York accent is widely recognized in the United States and, like many urban accents, evokes mainly negative reactions. One of the questions addressed in the sociolinguistic studies conducted in New York is that of rhoticity. It seems clear that the accent has made the historical transition from rhotic to non-rhotic, since it has (most of) the set of centring diphthongs ending in schwa which are characteristic of nonrhotic accents. However, there is considerable sociolinguistic variation with respect to rhoticity, and it appears that [ɹ] in coda position is staging a comeback. Recall that the standard accent in the United States, General American, is rhotic, and that rhoticity is regarded as more prestigious than non-rhoticity.

13.2.1.2 纽约市元音

13.2.1.2 New York City Vowels

[ɜɪ]元音

The [ɜɪ] vowel

/ɜ:/ 音素的 [ɜɪ] 实现，如 [hɜɪd] ( herd )，被广泛认为是纽约语言的特征，通常被认为是“布鲁克林”的特征accent'（尽管，正如我们之前提到的，它绝不仅限于布鲁克林）。然而，它被高度污名化并且可能正在消亡。这种认识发生在尾声辅音之前，因此它不会出现在purr的非发音发音中。一些说话者也有 /ɔɪ/ 音素的 [ɜɪ] 实现，也在尾辅音之前，因此一些最小对已成为同音字，如语音和诗歌。

The [ɜɪ] realization of the /ɜ:/ phoneme, as in [hɜɪd] (herd), is widely regarded as characteristic of New York speech, and is often said to characterize ‘the Brooklyn accent’ (although, as we noted earlier, it is by no means restricted to Brooklyn). However, it is highly stigmatized and is probably dying out. This realization occurs before a coda consonant, so that it does not occur in non-rhotic pronunciations of purr. Some speakers also have an [ɜɪ] realization of the /ɔɪ/ phoneme, also before a coda consonant, so that some minimal pairs have become homophones, as in voice and verse.

/æ/ 音位的变位音

Allophones of the /æ/ phoneme

在某些环境中，/æ/ 有可变的 [ε:ə] 和 [æ:] 实现，即在清擦音、浊塞音或鼻音出现在词尾尾音之前（尽管 /ŋ/ 表现可变），如在hash、past、bad、stabs、man和damp中，但在pal中则不同，末尾 /l/ 或hat末尾清音停顿。这些双元音或长实现有时被称为音素的“紧张”实现。目前尚不清楚为什么该位置的这些特定辅音会触发这种紧张过程。^3个变位音素的确切语音形式存在相当大的社会语言学差异，并且对于某些说话者，[ε:ə] 实现与 /εə/ 音素的实现合并，折叠最小对，例如bad/bared。

There are variable [ε:ə] and [æ:] realizations of /æ/ in certain environments, namely before a voiceless fricative, voiced stop or nasal when they occur in a word-final coda (although /ŋ/ behaves variably), as in hash, past, bad, stabs, man and damp, but not in pal, with a final /l/, or hat, with a final voiceless stop. These diphthongal or long realizations are sometimes referred to as ‘tense’ realizations of the phoneme. It is not clear why these specific consonants in that position should trigger this tensing process.3 There is considerable sociolinguistic variation in the exact phonetic form of the allophones, and, for some speakers, the [ε:ə] realization merges with realizations of the /εə/ phoneme, collapsing minimal pairs such as bad/bared.

/ɔ/的实现

Realizations of /ɔ/

说纽约的人通常对 /ɔ/ 音素有 [ɔə] 和 [oə] 的实现，如 [pʰɔə] ( paw /pour/pore )，以及起点更高的 [ʊə]，从而创造了可能性与 /ʊə/ 音素部分合并。

New York speakers often have [ɔə] and [oə] realizations of the /ɔ/ phoneme, as in [pʰɔə] (paw/pour/pore), and, with an even higher starting point, [ʊə], thus creating the possibility of partial merger with the /ʊə/ phoneme.

13.2.1.3 纽约市辅音

13.2.1.3 New York City Consonants

流变性和非流变性

Rhoticity and non-rhoticity

上面的讨论表明，纽约的演讲经历了从声响到非声响的转变，并且对于许多演讲者来说正在恢复到声响。这种现象很可能会导致更多的侵入性 [ɹ] 发生，因为一直没有发声但试图发声的说话者很可能会插入侵入性 [ɹ] 作为发出 [ɹ] 的一般策略的一部分，其中否则它可能不存在。

The discussion above shows that New York speech has undergone the transition from rhoticity to non-rhoticity, and is reverting, for many speakers, to rhoticity. This phenomenon may well be resulting in greater occurrence of intrusive [ɹ], since a speaker who has been non-rhotic but is attempting to be rhotic may well insert intrusive [ɹ]s as part of a general strategy to utter [ɹ] where it may otherwise be absent.

/θ/ 和 /ð/ 的实现

Realizations of /θ/ and /ð/

音位 /θ/ 和 /ð/ 通常实现为塞擦音（[t̪θ] 和 [d̪ð]）或牙塞音：[t̪] 和 [d̪]；这种变化是由社会语言学决定的。请注意，对于有牙塞的说话者，牙槽塞和牙塞之间的音位区别保持不变，如 [t̪ʰɪn] ( thin ) 与 [tʰɪn] ( tin )。对于许多说其他英语种类的人来说，这种区别可能很难注意到。

The phonemes /θ/ and /ð/ are often realized as either affricates ([t̪θ] and [d̪ð]) or dental stops: [t̪] and [d̪]; the variation is sociolinguistically determined. Note that, for speakers who have dental stops, the phonemic distinction between alveolar and dental stops is maintained, as in [t̪ʰɪn] (thin) vs [tʰɪn] (tin). To many speakers of other varieties of English, the distinction may well be difficult to notice.

/t/ 的实现

Realizations of /t/

与一般美国人一样，纽约语音在元音间环境（重读第一个元音）中有 /t/ 的颤动，但它在尾音位置也有比 GA 更大范围内 /t/ 的声门塞音实现。/t/ 音素经常被严重送气，有时在音节首位有擦伤音，如 [tsɪn] ( tin )。

New York speech, like General American, has Flapping of /t/ in intervocalic environments (where the first vowel is stressed), but it also has glottal stop realizations of /t/ in coda position on a greater scale than in GA. The /t/ phoneme is often heavily aspirated, to the point, at times, of being affricated, in syllable-initial position, as in [tsɪn] (tin).

gueqrecard_002

13.2.2 德克萨斯英语（曲目 13.7和练习 4）

13.2.2 Texan English (Track 13.7 and exercise 4)

13.2.2.1 定义德克萨斯口音

13.2.2.1 Defining the Texan Accent

得克萨斯州幅员辽阔，据说比法国还大。因此，德克萨斯英语中存在很多差异也就不足为奇了。许多受过教育的德州人会说德州英语和通用美语，具体取决于说话的语境。得克萨斯英语在该州的许多地方都存在风潮，但是，当人们接近与路易斯安那州的东部边界时，可以发现可变的非风潮。德克萨斯英语的变体是否可以被认为是美国英语的南方变体取决于人们认为什么构成了美国南部的语言：一些方言地图将德克萨斯排除在南部之外，而其他方言地图则包括该州的东部地区。人们在德克萨斯州遇到的许多发音特征都可以在邻近的州找到，但有一些发音特征据说是德克萨斯特有的。

Texas is a vast state, said to be larger than France. It is therefore unsurprising that there is much variability within Texan English. Many educated Texans speak both a Texan variety of English and General American, depending on the context of utterance. Texan English is rhotic in many parts of the state, but, as one approaches the eastern border with Louisiana, variable non-rhoticity can be found. Whether varieties of Texan English can be considered Southern varieties of US English depends on what one considers to constitute the linguistic American South: some dialect maps exclude Texas from the South, while others include eastern parts of the state. Many of the pronunciation features one encounters in Texas can be found in neighbouring states, but there are some pronunciation features that are said to be specifically Texan.

13.2.2.2 德克萨斯元音

13.2.2.2 Texan Vowels

词汇集 PRICE 的 /aɪ/ 音素在德克萨斯语中经常被单音化。这种现象被 Wells (1982) 非正式地称为 PRICE smoothing，导致长单元音，如price发音为 [pʰɹa:s]。在邻国可以听到价格平滑。

The /aɪ/ phoneme of the lexical set PRICE is often monophthongized in Texan speech. The phenomenon, referred to informally by Wells (1982) as PRICE smoothing, results in a long monophthong, as in price pronounced [pʰɹa:s]. PRICE smoothing can be heard in neighbouring states.

在词汇集 CHOICE 的词中 /ɔɪ/ 音素的德克萨斯语实现中也可以听到单音化，因此像oil这样的词发音为 [ɔ:l]，尽管音素有时会以 schwa off-glide 发音，如 [ɔɔl]。

Monophthongization can also be heard in the Texan realization of the /ɔɪ/ phoneme in words of the lexical set CHOICE, so that words such as oil are pronounced [ɔ:l], though the phoneme is sometimes pronounced with a schwa off-glide, as in [ɔɔl].

除了单元音化，在得克萨斯州和许多南部州，还有音素 /ɪ/、/ε/ 和 /æ/ 的双元音化，在词汇集 KIT、DRESS 和 TRAP 的单词中。所讨论的双元音有一个 schwa off-glide，因此kit发音为 [kʰɪət]，dress发音为 [dɹεəs]，trap发音为 [təɹæəp]。有时，人们可以听到 TRAP 组单词的发音是三元音或双音节发音，如 [bæjənd] 中的band。如果这样的发音有两个音节，那么我们可能会争辩说它们在第二个音节中包含一个 [æj] 双元音，后跟一个 schwa。

In addition to monophthongization, in Texas and in many Southern states, there is diphthongization of the phonemes /ɪ/, /ε/ and /æ/, in words of the lexical sets KIT, DRESS and TRAP. The diphthongs in question have a schwa off-glide, so that kit is pronounced [kʰɪət], dress is pronounced [dɹεəs] and trap is pronounced [təɹæəp]. One can, at times, hear pronunciations of words of the TRAP set with what is either a triphthong or a bisyllabic pronunciation, as in [bæjənd] for band. If such pronunciations have two syllables, then we might argue that they contain an [æj] diphthong followed by a schwa in the second syllable.

词汇集 DRESS 的单词中的 /ε/ 音素在音素 /ʃ/、/ʒ/ 和 /ŋ/ 之前也经过双元音化为 [ei]，因此单词special发音为 [speiʃəl]，听起来就像单词空间。

The /ε/ phoneme in words of the lexical set DRESS also undergoes diphthongization to [ei] before the phonemes /ʃ/, /ʒ/ and /ŋ/, so that the word special, pronounced [speiʃəl], sounds just like the word spatial.

在许多德州人的语音中，词汇集 DRESS 的 /ε/ 和词汇集 KIT 的 /ɪ/ 之间的音位区别在结尾 /n/ 之前被中和，因此Ken和kin等对都发音为 [ kʰɪn]。正如我们所见，中和被定义为特定语音环境中音位对比的中止（此处，在尾声 /n/ 之前）：表现出这种中和的说话者并不是失去了 /ε/ vs /ɪ/ 完全对比。

The phonemic distinction between the /ε/ of the lexical set DRESS and the /ɪ/ of the lexical set KIT is neutralized before coda /n/ in the speech of many Texans, so that pairs such as Ken and kin are both pronounced as [kʰɪn]. Neutralization is, as we have seen, defined as the suspension of a phonemic contrast in specific phono-logical environments (here, before coda /n/): it is not the case that speakers who exhibit this neutralization have lost the /ε/ vs /ɪ/ contrast altogether.

因此，/ε/ 音位在德克萨斯英语使用者中有多种实现，从 [εə] 到 [ei]，再到 [ɑ]。

So, the /ε/ phoneme has a wide variety of realizations among speakers of Texan English, from [εə], to [ei], to [ɑ].

词组MOTH中的/aʊ/双元音在得克萨斯州和邻近州往往起点较高，所以mouth这个词发音为[mεʊθ]。

The /aʊ/ diphthong in words of the set MOTH often has a higher starting point, in Texas and in neighbouring states, so that the word mouth is pronounced [mεʊθ].

词汇集 LOT 的单词中的音素在德克萨斯英语中通常具有双元音实现，因此单词dog发音为 [doʊg]，通常非正式地写为dawg。鉴于此和 MOUTH 元音，复合词如cow dogs（放牛的狗）可以发音为 [kʰεʊdoʊgz]。

The phoneme in words of the lexical set LOT often has a diphthongal realization in Texan English, so that the word dog is pronounced [doʊg], often written informally as dawg. Given this and the MOUTH vowel, a compound such as cow dogs (dogs for herding cows) can be pronounced as [kʰεʊdoʊgz].

在德克萨斯英语中 /u:/ 与 /ʊ/ 的区别通常在结尾 /l/ 之前被中和，因此pull和pool都发音为 [pʰʊl]。

The /u:/ vs /ʊ/ distinction is often neutralized before coda /l/ in Texan English, so that pull and pool are both pronounced [pʰʊl].

13.2.2.3 德克萨斯辅音

13.2.2.3 Texan Consonants

德克萨斯英语的辅音与普通美国英语的辅音大致相似，在所有上下文中都具有“dark l”、Flapping 和 rhoticity。然而，当人们接近路易斯安那州的边界时，可以观察到可变的非流变性，因为德克萨斯州东部的南方口音基本上是非流变性的。

The consonants of Texan English are broadly parallel to those of General American, featuring ‘dark l’ in all contexts, Flapping and rhoticity. However, as one approaches the border with Louisiana, variable non-rhoticity can be observed, since the Southern accents to the east of Texas are largely non-rhotic.

我们看到 SSE 说话者保留了 /w/ 与 /ʍ/ 成对的区别，例如witch/which和Wales/whales。这也可以在一些说德克萨斯英语的人中得到证实。

We saw that SSE speakers have retained the /w/ vs /ʍ/ distinction in pairs such as witch/which and Wales/whales. This can also be attested among some speakers of Texan English.

13.3 两种南半球口音

13.3 Two Southern Hemisphere Accents

gueqrecard_002

13.3.1 澳大利亚英语（曲目 13.8和练习 5）

13.3.1 Australian English (Track 13.8 and exercise 5)

13.3.1.1 定义澳大利亚英语口音

13.3.1.1 Defining the Australian English Accent

对澳大利亚英语的描述通常区分三种社会定义的变体：Cultivated、General 和 Broad Australian。我们不检查它们之间的差异，这些差异主要涉及元音发音。但是，我们将简要概述整个澳大利亚都在使用的通用澳大利亚语。澳大利亚英语的通用发音起源于 19 世纪早期英格兰东南部工人阶级的讲话，因此在某些方面类似于当今的伦敦英语。

Descriptions of Australian English often distinguish between three socially defined varieties: Cultivated, General and Broad Australian. We do not examine the differences between these, which mostly concern vowel articulations. We will, however, give a brief overview of General Australian, which is spoken throughout Australia. General Australian English pronunciation has its origins in the speech of early nineteenth-century working-class speakers from the South East of England, and is therefore similar to present-day London English in some respects.

13.3.1.2 澳大利亚通用元音

13.3.1.2 General Australian Vowels

一般澳大利亚语的主要特征可以在元音发音中找到。通用澳大利亚语的元音系统与 RP 的元音系统相似，但有许多主要的实现差异，主要是“元音转换”性质，正如我们在第 12 章前面提到的那样。

The main characterizing properties of General Australian are to be found in vowel articulations. The vowel system of General Australian is parallel to that of RP, but with many major realizational differences, mostly of a ‘vowel shift’ nature, as we noted earlier in chapter 12.

/i:/, /eɪ/, /aɪ/, /ɔɪ/ 元音转换

The /i:/, /eɪ/, /aɪ/, /ɔɪ/ vowel shift

我们在第 12 章描述了这种元音转换。它可以描述如下：

We described this vowel shift in chapter 12. It can be depicted as follows:

/u:/, /oʊ/, /aʊ/ 元音转换

The /u:/, /oʊ/, /aʊ/ vowel shift

与高前不圆唇音位 /i:/ 一样，高后圆唇元音音位 /u:/ 已经双元音化，通常导致双元音具有高前部不圆唇起点和高后部不圆唇结束点；我们将其转录为 [ɪɯ]。这种实现潜在地侵占了 /oʊ/ 的 [əʊ] 类型实现的空间，后者已经转移到 [ɐʉ]，具有较低的中心不圆起点和较高的中心圆终点，从而进入 /aʊ 的空间/，其实现转移到[æo]，起点比[ɐʉ]更靠前，终点更低更靠后。这组移位可以描述如下：

Like the high front unrounded phoneme /i:/, the high back rounded vowel phoneme /u:/ has diphthongized, often resulting in a diphthong with a high front unrounded starting point and a high back unrounded finishing point; we will transcribe this as [ɪɯ]. This realization potentially encroaches on the space of the [əʊ]-type realizations of /oʊ/, which have shifted to [ɐʉ], with a low central unrounded starting point and a high central rounded end point, thus entering the space of /aʊ/, whose realization has shifted to [æo], in which the starting point is more front than that of [ɐʉ], and the end point lower and further back. This set of shifts can be depicted as follows:

/ʌ/, /æ/, /ε/, /ɪ/ 元音转换

The /ʌ/, /æ/, /ε/, /ɪ/ vowel shift

元音移位也影响了短元音 /ʌ/、/æ/、/ε/ 和 /ɪ/，其中 /ʌ/ 在 [a] 区域被实现为低前发音，靠近 / 的空间æ/ 音素，实现为 [ε]-like。反过来，/ε/ 被实现为 [e]-like，因此接近 /ɪ/ 的空间，后者更像是 [i]-like。当然，这种发音不同于 /i:/，正如我们所见，后者是双元音。我们可以将这组元音移位描述如下：

A vowel shift has also affected the short vowels /ʌ/, /æ/, /ε/ and /ɪ/, with /ʌ/ being realized as a low front articulation in the [a] area, close to the space of the /æ/ phoneme, which is realized as [ε]-like. In turn, /ε/ is realized as [e]-like, and thus close to the space of /ɪ/, which is rather [i]-like. This articulation is, of course, distinct from that of /i:/, which, as we have seen, is diphthongal. We may depict this set of vowel shifts as follows:

/ɑ:/ 与 /ʌ/ 的区别

The /ɑ:/ vs /ʌ/ distinction

/ɑ:/ 的通用澳大利亚语实现，与 /ʌ/ 一样，也位于前面，但两者之间的区别并未合并，因为它们之间存在长度差异，如 [Pʰat] ( putt ) 与 [ Pʰa:t]（部分）。

The General Australian realization of /ɑ:/, like that of /ʌ/, is also fronted, but the distinction between the two is not merged, since there is a length difference between them, as in [Pʰat] (putt) vs [Pʰa:t] (part).

13.3.1.3 通用澳大利亚辅音

13.3.1.3 General Australian Consonants

我们已经注意到，General Australian 和 SSE 一样，在所有位置都有“dark l”；在这两种情况下，“黑暗”的确切性质可能需要舌体的收缩和降低，而不是这样的软化。

We have already noted that General Australian, like SSE, has ‘dark l’ in all positions; the precise nature of the ‘darkness’ may entail, in both cases, retraction and lowering of the tongue body, rather than velarization as such.

一般澳大利亚人是非风湿性的。它也有一个与北美英语中的 Flapping 非常相似的过程：a /t/ 通常会被理解为元音之间的浊音。

General Australian is non-rhotic. It also has a process rather similar to that of Flapping in North American English: a /t/ will often be realized as a voiced articulation between vowels.

gueqrecard_002

13.3.2 印度英语（曲目 13.9和练习 6）

13.3.2 Indian English (Track 13.9 and exercise 6)

13.3.2.1 定义印度英语

13.3.2.1 Defining Indian English

印度次大陆使用多种语言。它们分为两个语系。北方的语言属于印欧语系语族。这是一个庞大的历史相关语言家族。在 19 世纪，语言学家能够证明印度北部的一种古老语言梵语最终与古希腊语和拉丁语有关。从这一发现中，可以证明印度北部的语言与欧洲使用的许多语言有关，例如罗曼语（包括法语、西班牙语、葡萄牙语和意大利语）和日耳曼语（包括英语）、德语、荷兰语和瑞典语）。印度北部的印欧语系包括印地语、马拉地语、古吉拉特语和旁遮普语。很难相信，当听到这些语言时，它们在历史上与英语相关，但它们确实如此；必须记住，这些语言进化的时间跨度很大。印度南部使用的语言属于一个完全不同的语系，即德拉威语系。其中包括泰米尔语、马拉雅拉姆语、卡纳达语、泰卢固语和托达语。印度的语言，无论是印欧语还是德拉威语，都无法相互理解。英语因此具有了一种地位印度的通用语：一种可以用作母语无法相互理解的人之间交流的语言。许多受过教育的印度人将英语作为第二语言，因此使用英语作为通用语。

There are many languages spoken on the Indian subcontinent. They are divided into two language families. The languages of the North belong to the Indo-European language family. This is a vast family of historically related languages. In the nineteenth century, linguists were able to show that Sanskrit, an ancient language of the North of India, was ultimately related to Ancient Greek and to Latin. From this discovery, it was possible to show that the languages of the North of India were related to many of the languages spoken in Europe, such as the Romance languages (including French, Spanish, Portuguese and Italian) and the Germanic languages (including English, German, Dutch and Swedish). The Indo-European languages of the North of India include Hindi, Marathi, Gujerati and Punjabi. It is hard to believe, when listening to those languages, that they are historically related to English, but they are; it is has to be borne in mind that the time-span over which these languages have evolved is vast. The languages spoken in the South of India belong to an entirely distinct language family, known as the Dravidian language family. These include Tamil, Malayalam, Kannada, Telugu and Toda. The languages of India, whether Indo-European or Dravidian, are not mutually comprehensible. English has therefore taken on the status of a lingua franca in India: a language which can be used as a form of communication between people whose native languages are not mutually comprehensible. Many educated Indians have English as a second language, and thus use English as a lingua franca.

13.3.2.2 印度英语元音

13.3.2.2 Indian English Vowels

集合 FACE 和 GOAT 的元音在 RP 和 GA 中是双元音，在印度英语中通常实现为单元音 [e] 和 [o]，因此 FACE 是 [fes]，GOAT 是 [got]。

The vowels of the sets FACE and GOAT, which are diphthongs in RP and GA, are typically realized as the monophthongs [e] and [o] in Indian English, so that FACE is [fes] and GOAT is [got].

许多（但不是全部）讲印度英语的人都是变音的，这对元音系统有影响：正如我们所见，在变音的重音中没有 /ɪə/、/εə/、/ʊə/英语。词汇集 NURSE 单词中的 /ɜ:/ 元音通常不存在于印度英语的变体或非变体中。即使在讲印度英语的人不会流音的情况下，也可以在 NURSE 词集的单词中找到词集 STRUT 中的 [ʌ] 元音，因此诸如merlot（源自法语的英语单词，表示用于生产同名红酒的葡萄），在 RP 中发音为 [ˈmɜ:ləʊ]，发音为 [ˈmΛlo]。

Many, but not all, speakers of Indian English are rhotic, and this has consequences for the vowel system: as we have seen, there are no centring diphthongs of the sort /ɪə/, /εə/, /ʊə/ in rhotic accents of English. The /ɜ:/ vowel in words of the lexical set NURSE is often not present in either rhotic or non-rhotic varieties of Indian English. Even among non-rhotic speakers of Indian English, a vowel of the sort [ʌ], of the lexical set STRUT, can be found in words of the NURSE set, so that words such as merlot (an English word of French origin, denoting a grape used to produce a red wine of that name), pronounced [ˈmɜ:ləʊ] in RP, is pronounced [ˈmΛlo].

长元音 /ɔ:/（词汇集 THOUGHT、NORTH 和 FORCE）和短元音 /ɔ:/（词汇集 LOT）之间的 RP 区别在各种印度英语中通常不存在。RP 最小对，例如awful ([ˈɔ:fəə]) 和offal ([ˈɒfəə]) 因此在许多讲印度英语的人的语音中是同音字，两者都发音为 [ˈɔfəə]，带有短 [ɔ]，如文字游戏话语“英国厨房里有大量内脏被扔掉。”

The RP distinction between the long vowel /ɔ:/ (of the lexical sets THOUGHT, NORTH and FORCE) and the short vowel /ɔ:/ (of the lexical set LOT) is often absent in varieties of Indian English. RP minimal pairs such as awful ([ˈɔ:fəɫ]) and offal ([ˈɒfəɫ]) are therefore homophones in the speech of many speakers of Indian English, both being pronounced [ˈɔfəɫ], with a short [ɔ], as in the word-play utterance ‘There’s an awful lot of offal being thrown away in British kitchens.’

印度英语有 happY Tensing，所以happy这个词发音为 [hæpi]，就像在 RP 中一样。

Indian English has happY Tensing, so that the word happy is pronounced [hæpi], as in RP.

13.3.2.3 印度英语辅音

13.3.2.3 Indian English Consonants

印度北部的一些印欧语系在清不送气塞音、清送气塞音、浊塞音之间有四向音位对比停止和呼吸声停止。我们已经在spit、stick和skin等英语单词中遇到过清音不送气塞音。我们也遇到过英语单词中的送气清塞音，例如pit中的 [pʰ] ；tip和king中的词首塞音也送气。浊音出现在labour、ladder和wriggle等词中. 我们还没有遇到有呼吸声的停止。在浊音中，声带靠在一起振动。在呼吸浊音中，声带分开得更远，但由于通过声门的气流增加，声带仍然振动。双唇、肺泡和软腭呼吸浊塞音被转录为 [b̤]、[d̤] 和 [ɡ̤]。它们出现在印度烹饪中的印地语单词中，例如bhindi（秋葵）、dhania（芫荽籽）和ghee（澄清黄油）。具有这种四向对比的最小对的示例是 [bal]（“头发”）vs [pal]（“照顾”）vs [p̤al]（“刀片”）vs [b̤al]（“额头”）。您可以在音轨 13.1的其他发音位置听到这些和最小对。

Some of the Indo-European languages of the North of India have a four-way phonemic contrast between voiceless unaspirated stops, voiceless aspirated stops, voiced stops and breathy voiced stops. We have already encountered voiceless unaspirated stops in English words such as spit, stick and skin. We have also encountered aspirated voiceless stops in English words such as the [pʰ] in pit; the initial stops in the words tip and king are also aspirated. Voiced stops occur in words such as labour, ladder and wriggle. We have not yet encountered breathy voiced stops. In voiced sounds, the vocal cords are vibrating close together. In breathy voiced stops, the vocal cords are further apart, but still vibrate because of increased airflow through the glottis. The bilabial, alveolar and velar breathy voiced stops are transcribed as [b̤], [d̤] and [ɡ̤]. They occur in Hindi words which feature in Indian cookery, such as bhindi (okra), dhania (coriander seeds) and ghee (clarified butter). Examples of minimal pairs featuring this four-way contrast are [bal] (‘hair’) vs [pal] (‘take care of’) vs [p̤al] (‘knife blade’) vs [b̤al] (‘forehead’). You can hear these, and minimal pairs at other places of articulation, on Track 13.1.

gueqrecard_002

说印度英语的人通常会说出诸如pit、tip和king等在初始位置带有不送气清塞音的词。从表面上看，这很令人费解，因为许多讲印度英语的人在他们的印度母语中都有一系列送气清塞音：为什么在说pit、tip和king等词时不直接使用母语的送气清塞音？如果我们考虑到许多印度本土语言缺乏牙擦音，就可能对这种现象有所了解；因此，讲印度英语的人会进行“TH-stopping”，他们会为 /θ/ 发出送气清齿音 [t̪̪] 并为 /ð/ 发出带呼吸的浊齿音 [d̪]。可能是因为送气清塞音系列 ([tʰ]) 的一个成员被用于 /θ/，所以不送气系列被用在原本在 RP 中会有送气清塞音的单词中。

It is common for speakers of Indian English to produce words such as pit, tip and king with unaspirated voiceless stops in initial position. On the face of it, this is puzzling, since many speakers of Indian English have a series of aspirated voiceless stops in their native Indian language: why not just use the aspirated voiceless stops of one’s native language when uttering words like pit, tip and king? Some insight into this phenomenon may be gained if we consider that many of the native Indian languages lack dental fricatives; speakers of Indian English therefore engage in ‘TH-stopping’, in which they produce aspirated voiceless dental [t̪̪] for /θ/ and the breathy voiced dental stop [d̪] for /ð/. Perhaps it is because a member of the aspirated voiceless stop series ([tʰ]) is being used for /θ/ that the unaspirated series is used in words which would otherwise have an aspirated voiceless stop in RP.

印度的许多母语都有一系列的卷舌塞音。舌尖和舌叶向后卷曲时产生卷舌音，因此舌头的下侧与牙槽嵴形成发音。以这种方式形成的塞音包括清卷舌塞音 [ʈ] 和浊卷舌塞音 [ʈ]。说印度英语的人经常以这种卷舌方式来理解 /t/ 和 /d/；这经常可以在印度餐馆服务员的发音中听到，他们经常会说出带有卷舌 [ʈ] 的chapati（一种印度大饼）和带有卷舌 [ɖ] 的poppadum（一种薄而脆的印度面包）等词. 您可以在曲目 13.2上听到这个。

Many of the native languages of India have a series of retroflex stops. Retroflex sounds are produced with the tip and blade of the tongue curled backwards, so that the underside of the tongue forms an articulation with the alveolar ridge. Stops formed this way include the voiceless retroflex stop [ʈ] and the voiced retroflex stop [ʈ]. Speakers of Indian English often realize /t/ and /d/ in this retroflex manner; this can often be heard in the pronunciation of waiters in Indian restaurants, who will often produce words such as chapati (an Indian flatbread) with a retroflex [ʈ], and poppadum (a thin, crispy Indian bread) with a retroflex [ɖ]. You can hear this on Track 13.2.

在大多数印度英语中，/l/ 音素并未被理解为“暗”（软化）侧音。/r/ 音素通常被理解为肺泡拍打声，因此rap一词发音为 [ɾap]。音素 /v/ 和 /w/ 并不总是被讲印度英语的人一贯区分。印度的许多本土语言都有一个介于近似音 [w] 和摩擦音 [v] 之间的近似音。他们发出的声音是唇齿合音，其中下唇与上牙形成开放式合音的狭窄。这个声音被转录为 [υ]。它像 [w] 因为它是一个近似值，它像 [v] 因为它是唇齿音。您可以在Track 13.3中听到 [v]、[υ] 和 [w] 之间的对比（插图来自非洲语言 Isoko）. 说印度英语的人，因为他们没有 /v/ 和 /w/ 的区别，所以通常会用 [w] 来表示 [v]，或者用 [υ] 来表示 [v]。

The /l/ phoneme is not realized as a ‘dark’ (velarized) lateral in most Indian English. The /r/ phoneme is often realized as an alveolar tap, so that the word rap is pronounced as [ɾap]. The phonemes /v/ and /w/ are not always consistently distinguished by speakers of Indian English. Many of the native languages of India have an approximant which is intermediate between the approximant [w] and the fricative [v]. The sound they have is a labio-dental approximant, in which the lower lip forms a stricture of open approximation with the upper teeth. This sound is transcribed as [υ]. It is like a [w] in that it is an approximant, and it is like a [v] in that it is labio-dental. You can hear the contrast between [v], [υ] and [w] (illustrated from the African language Isoko) on Track 13.3. Speakers of Indian English, because they lack a /v/ vs /w/ distinction, will often produce [w] for [v], or a [υ] for [v].

gueqrecard_002

印度的许多母语都没有重音计时，因此印度英语通常没有重音计时，而许多英语变体都是典型的重音计时。与此相关的是，在印度英语中，stress 这个词经常出现在错误的音节上。

Many of the native languages of India are not stress-timed, so Indian English is often spoken without the stress timing that is typical of so many varieties of English. Linked to this is the fact that the word stress often falls on the wrong syllable in Indian English.

13.4 口音变异中的一些常见现象概述

13.4 An Overview of Some Common Phenomena Found in Accent Variation

从我们对上述各种英语口语的简要描述中可以清楚地看出，任何语音社区的语音，实际上是社区内给定说话者的语音，通常都是可变的。随着时间的推移，正是这种差异最终会导致不同社区的言论出现分歧。这种变化涉及几个不同的因素。许多社会语言学因素都是相关的。这些可能包括说话者的性别、年龄或社会阶层、说话者的社会抱负、说话者所处社会的结构以及说话者所处社会网络的复杂方面，涉及诸如团结、群体身份和个人身份。我们没有在这里检查这些因素，

It is clear from our brief description of the above varieties of spoken English that the speech of any speech community, and indeed of a given speaker within a community, is typically variable. It is such variation that can eventually lead to divergence between the speech of different communities over time. Several different factors are involved in such variation. A host of sociolinguistic factors is relevant. These may include the sex, age or social class of the speaker, the speaker’s social aspirations, the structure of the society in which the speaker lives, and complex aspects of the social networks in which the speaker lives, involving such things as solidarity, group identity and individual identity. We have not examined these factors here, but we note that an understanding of them is vital if we are to gain a comprehensive understanding of variation in pronunciation in a given speech community.

发音的变化也受到社会因素以外的因素的限制，例如声道的性质、音素系统中音素之间的关系，以及与声音的相对感知显着性有关的因素，这取决于它们是什么声音之前或之后，它们出现在音节结构中的位置，以及它们是否出现在重读或非重读音节中。以下是我们在描述口音变化时讨论过的各种现象的总结。

Variation in pronunciation is also constrained by factors other than societal ones, such as the nature of the vocal tract, the relationship between the phonemes within a phoneme system, and factors to do with the relative perceptual salience of sounds, depending on what sounds they are preceded or followed by, where they occur within syllable structure, and whether they occur in stressed or unstressed syllables. The following is a summing up of the sorts of phenomena we have discussed in describing accent variation.

13.4.1元音现象

13.4.1 Vowel Phenomena

13.4.1.1 双元音化

13.4.1.1 Diphthongization

元音音素的双元音实现可能由相邻的辅音触发，如 /i:/ 在“dark l”之前的 [i:ə] 实现，或者可能“自发地”发生；我们在通用美国英语、纽约城市英语、通用澳大利亚英语、伦敦英语、RP 和泰恩赛德英语中看到了后者的许多例子。同样，RP 曾经有中元音 /e/ 和 /o/ 的单元音实现，但这些在 RP 的历史上经历了双元音化。这些中元音的双元音化在 GA 的历史中也发生过，但在 SSE 中没有。

Diphthongal realizations of vowel phonemes may be triggered by an adjacent consonant, as in the [i:ə] realization of /i:/ before ‘dark l’, or may occur ‘spontaneously’; we have seen many examples of the latter in General American, New York City English, General Australian, London English, RP and Tyneside English. Similarly, RP used to have monophthongal realizations of the mid vowels /e/ and /o/, but these have undergone diphthongization in the history of RP. Diphthongization of these mid vowels has also taken place in the history of GA, but not in SSE.

13.4.1.2 单体化

13.4.1.2 Monophthongization

我们已经看到许多双元音的单元音实现的例子，例如 [ε:] 代替 [εə] 和 [ɔ:] 代替 RP 中的 [ʊə]。

We have seen many examples of monophthongal realizations of diphthongs, as in [ε:] instead of [εə] and [ɔ:] instead of [ʊə] in RP.

13.4.1.3 拆分

13.4.1.3 Splits

正如我们所见，音位可能有音位变体，但如果互补分布模式被打乱，这些音位变体可能会获得音位地位，从而导致新出现的平行分布模式。示例包括 /ə/ 结尾的居中双元音音位（/ɪə/、/εə/、/ʊə/）的出现，以及 RP 和其他重音中的 /ʊ/ 与 /ʌ/ 区别（FOOT/STRUT 拆分）英语的。

As we have seen, phonemes may have allophones, but those allophones may gain phonemic status if a pattern of complementary distribution is disrupted, resulting in a newly emergent pattern of parallel distribution. Examples include the emergence of the /ə/-ending centring diphthong phonemes (/ɪə/, /εə/, /ʊə/), and the /ʊ/ vs /ʌ/ distinction (the FOOT/STRUT split) in RP and other accents of English.

13.4.1.4 合并

13.4.1.4 Mergers

由于发音可能会发生变化，因此一个音素的实现可能会与另一个音素的实现合并，从而导致音位区别的丧失。例如，在许多英语口音中 /k/ 与 /x/ 和 /ʍ/ 与 /w/ 对比的合并，导致以前最小对的词变成同音词，如lock/loch和which/witch。

Since articulations may shift, it is possible for the realizations of one phoneme to merge with those of another, resulting in the loss of a phonemic distinction. Examples are the mergers of the /k/ vs /x/ and /ʍ/ vs /w/ contrasts in many accents of English, resulting in words which were previously minimal pairs becoming homophones, as in lock/loch and which/witch.

13.4.1.5 元音移位

13.4.1.5 Vowel Shifts

当一个音素的实现侵占另一个元音音素的实现时，可能会采取“规避行动”，从而维持音素对比，避免合并。我们在伦敦英语和澳大利亚英语中都看到过这样的例子。

When the realization of one phoneme encroaches on the realization of another vowel phoneme, ‘evasive action’ may be taken, so that the phonemic contrast is sustained and merger is avoided. We have seen examples of this in both London English and Australian English.

13.4.1.6 减少元音

13.4.1.6 Vowel Reduction

元音经常缩减为弱形式（例如 /i:/ → [i] 和 /u:/ → [u]），或者缩减为 schwa，在非重读音节中，如我们所见，这些音节在感知上不那么突出。这种现象几乎出现在各种英语中。

Vowels frequently reduce to either a weak form (e.g. /i:/ → [i] and /u:/ → [u]), or to schwa, in unstressed syllables which, as we have seen, are perceptually less salient. This phenomenon occurs in almost every variety of English.

13.4.2 辅音现象

13.4.2 Consonantal Phenomena

13.4.2.1 辅音弱化

13.4.2.1 Weakening of Consonants

一个常见的现象是元音间弱化，其中辅音发音变得更像元音，在变得浊音的意义上，或者经历狭窄程度或持续时间的减少。北美英语中的拍打声、澳大利亚英语中 /t/ 的发音以及泰恩赛德英语中的“T-to-R”都是这种现象的例子。

A common phenomenon is intervocalic weakening, in which a consonant articulation becomes more vowel-like, in the sense of becoming voiced, or undergoing a diminution in degree or duration of stricture. Flapping in North American English, voicing of /t/ in Australian English and ‘T-to-R’ in Tyneside English are all examples of this phenomenon.

在尾声位置，辅音经常以狭窄程度降低的形式减弱，有时会导致完全省略。示例是非喉音口音中尾音 [ɹ] 的侵蚀和最终丢失，以及清塞音减少为声门塞音，这在某种程度上发生在我们考虑的所有口音中。出现在伦敦英语中的结尾 /l/ 的发声是另一个这样的过程。

In coda position, consonants often undergo weakening in the form of reduction in degree of stricture, sometimes leading to complete elision. Examples are the erosion and eventual loss of coda [ɹ] in non-rhotic accents, and the reduction of voiceless stops to glottal stops, which occurs to some extent in all of the accents we have considered. Vocalization of coda /l/, which occurs in London English, is another such process.

13.4.2.2 停止的摩擦

13.4.2.2 Affrication of Stops

这种现象，其中 /p/、/t/ 和 /k/ 可以分别实现为 [pΦ]/[pf]、[tθ]/[ts] 和 [kx]，似乎与强烈的愿望有关。它已在纽约市、利物浦和伦敦得到证实。

This phenomenon, in which /p/, /t/ and /k/ may be realized as [pΦ]/[pf], [tθ]/[ts] and [kx] respectively, appears to be connected with strong aspiration. It has been attested in New York City, Liverpool and London.

关于此类过程要牢记的要点是，它们很少局限于特定的英语口音。它们也不限于英语。因为它们源于人类声道的性质、人类感知能力和人类语言音系结构，所以它们在世界语言中得到广泛证实。

The principal point to be borne in mind about such processes is that they are rarely limited to specific accents of English. Nor are they limited to English. Because they arise from the nature of the human vocal tract, human perceptual capacities and the structure of human language phonologies, they are widely attested across the world’s languages.

笔记

Notes

1在这种情况下，元音通常有明显的鼻音化，鼻音跟在开元音之后。

1 There is often quite marked nasalization of the vowel in cases like these, where a nasal follows an open vowel.

2事实上，一些泰恩赛德人确实在 [ʊ] 和不圆润的、有时集中的 [ʊ] 之间形成对比，我们可以将其转录为 [ɨ]。

2 Some Tyneside speakers do in fact have a contrast, between [ʊ] and an unrounded, sometimes centralized, version of [ʊ], which we might transcribe as [ɨ].

3人们经常注意到，以紧尾辅音结尾的助动词（例如can和had）不会紧缩。

3 It has often been noted that auxiliary verbs which end with a tensing coda consonant, such as can and had, nevertheless do not undergo tensing.

练习

Exercises

本章末尾给出了练习的参考资料。

References for the exercises are given at the end of the chapter.

1 伦敦英语

听曲目 13.4。这是来自 IViE 语料库的声音文件（英语语调变化：有关详细信息和进一步参考，请参阅 Grabe、Nolan 和 Farrar 1998），其中包含来自伦敦、剑桥、布拉德福德、利兹、利物浦、纽卡斯尔、加的夫的英国人的录音，北爱尔兰（贝尔法斯特）和爱尔兰共和国（都柏林）。您可以在以下网站访问语料库：www.phon.ox.ac.uk/files/apps/IViE。或者，在搜索引擎上输入“IViE”。演讲者正在朗读以下段落的前两段：

从前，有一个女孩叫灰姑娘，但大家都叫她灰姑娘。Cinders 和她的母亲以及两个继姐妹 Lilly 和 Rosa 住在一起。莉莉和罗莎非常不友好，她们是懒惰的女孩。他们把所有的时间都花在买新衣服和参加派对上。可怜的 Cinders 不得不穿上他们所有的旧衣服！她还得打扫卫生！

Once upon a time, there was a girl called Cinderella, but everyone called her Cinders. Cinders lived with her mother and two stepsisters called Lilly and Rosa. Lilly and Rosa were very unfriendly and they were lazy girls. They spent all their time buying new clothes and going to parties. Poor Cinders had to wear all their old hand-me-downs! And she had to do the cleaning!

一天，一位皇家使者前来宣布舞会。舞会将在皇宫举行，以纪念女王的独子威廉王子。莉莉和罗莎认为这是神圣的。威廉王子很漂亮，他正在寻找新娘！他们梦想着结婚的钟声！

One day, a royal messenger came to announce a ball. The ball would be held at the Royal Palace, in honour of the queen’s only son, Prince William. Lilly and Rosa thought this was divine. Prince William was gorgeous, and he was looking for a bride! They dreamed of wedding bells!

舞会的傍晚到来时，辛德斯不得不帮助她的姐妹们做好准备。他们心情不好。他们想买几件新礼服，但他们的母亲说他们有足够的礼服。所以他们开始对着 Cinders 大喊大叫。'找到我的珠宝！一个叫道。“找到我的帽子！” 另一个嚎叫。他们想要发刷、发夹和发胶。

When the evening of the ball arrived, Cinders had to help her sisters get ready. They were in a bad mood. They’d wanted to buy some new gowns, but their mother said that they had enough gowns. So they started shouting at Cinders. ‘Find my jewels!’ yelled one. ‘Find my hat!’ howled the other. They wanted hairbrushes, hairpins and hairspray.

(a) 说话者是否始终不发脾气？

(b) 在 RP 中有词尾 schwa 的词中（例如Cinderella和Cinders），你发现这个说话者的元音是什么？

(d) 这个说话者表现出 /l/ 发声。我们已经说过，这发生在伦敦英语的音节押韵中，而不是出现在尾声中。这对这位演讲者来说是真的吗？

(e) 这个说话者在什么情况下产生 /t/ 音素的声门塞音实现？

2 泰恩赛德英语（纽卡斯尔）

gueqrecard_002

听曲目 13.5。这也是一个 IViE 文件。演讲者正在朗读练习 1 中灰姑娘段落的所有三段。

(a) 据说英格兰北部口音没有 FOOT/STRUT 拆分（即 /ʊ/ 和 /ʌ/ 之间的音位区别）。这位演讲者是否属于这种情况？提供音轨上相关单词的示例。

(b) 在以 <-er(s)> 结尾的单词中，例如Cinders和mother，说话者在最后一个非重读音节中发出的元音是什么？这是您对 Geordie 扬声器的期望吗？

(d) 在诸如clothes之类的词中，说话者是否有元音的单元音或双元音实现？

(e) 据说说 Geordie 的人在所有音节上下文中都“清晰”（颚化）[l ^{j ]。}这对这位演讲者来说是真的吗？

(f) 引用这位演讲者演讲中声门效应的例子。说话者将 /d/ 理解为一个词中的声门塞音：你能认出这是哪个词吗？

(g) 我们说过居中双元音 /ɪə/ 和 /ʊə/ 在泰恩赛德英语中通常实现为 /ɪɐ/ 和 /ʊɐ/。你能在录音中听到这方面的任何例子吗？

3 标准苏格兰英语（格拉斯哥）

gueqrecard_002

听曲目 13.6。这是来自 PAC 项目的录音（La Phonologie de l'anglais contemporain；有关详细信息，请参阅 Carr、Durand 和 Pukli 2004）。PAC 网站位于： http: //www.projet-pac.net/

演讲者正在阅读以下 PAC 单词列表：

1.坑

2.宠物

3.拍拍

4.锅

5.放

6.推杆

7.海

8.说

9.叹息

10.起诉

11.搅拌

12. 驾驭

13.楼梯

14. 犯错

15.远

16.战争

17. 更多

18.咕噜声

19. 沼泽地

20.感觉

21.填充

22.跌倒

23. 秋天

24.满

25.傻瓜

26.失败

27. 小马驹

28.档案

29.犯规

30.贴膜

31. 毛皮

32.鸟

33.吟游诗人

34. 胡须

35. 裸露

36.板

37.禁止

38.无聊

39.博德

40.鞠躬

41.珠子

42.出价

43.床

44.坏

45.吟游诗人

46. 裤子

47.植物

48.大师

49.之后

50.蚂蚁

51.阿姨们

52.跳舞

53.更远

54.父亲

55.排

56.玫瑰

57.行

58.毛孔

59.可怜

60.倒

61.爪子

62.爪子

63.暂停

64.姿势

65.等待

66.体重

67.边

68.叹了口气

69.同意

70.贪婪

71.育雏

72.酿造

73.冷杉

74.公平

75.毛皮

76.四

77. 前

78.为了

79.鼻子

80.知道

81.婴儿床

82.抓到

83.肉

84.见面

85. 伙伴

86.一无所获

87.结

88.娃娃

89. 救济金

90.凶

91.鸟

92.稀缺

93. 专家

94.开始

95.马

96.嘶哑

97.单词

98.葫芦

99.短

100.运动

101.下一个

102.烦恼

103.豹

104.牧羊人

105.这里

106.那里

107.疲倦

108.精神

109.结婚

110.玛丽

111.快活

112.对不起

113.故事

114.快点

115.陪审团

116.埋葬

117.浆果

118.天堂

119.面酵

120.地球

121.泊位

122.厨师

123.烟灰

124.看

125.房间

126.珍珠

127.危险

(a) 据说 SSE 是风动的，但也有一些演讲者是非风动的。这个演讲者一直很吵吗？

(b) 据说 SSE 没有长 /u:/ 和短 /ʊ/ 之间的对比，后者在 RP 中发现。相反，据说 SSE 有一个单一的 /u/ 音素，实现为高中央圆唇元音 [ʉ]。录音中的演讲者是否如此？

(c) 某些 SSE 使用者在bird、curd和heard等词中只有双向对比，bird和curd有 /ʌ/，heard有 /ε/。其他人有三向对比，在bird中使用 /ɪ/，在curd中使用/ʌ/，在heard中使用/ε/ 。这个扬声器在这些“pre-r”位置有双向对比还是三向对比？

(d) 据说 SSE 在coat中的短 /o/ 和cot中的短 /ɔ/之间具有对比。据说在 RP 中，词汇集 LOT（在 RP 中有一个短 /ɔ:/）的词与 THOUGHT 和 NORTH 词集（在 RP 中有一个长 /ɔ: / 在 RP)。该扬声器是否在诸如cot / caught和knot / naught之类的对之间表现出任何长度差异？

(e) 说话者通常是 SSE，因为她有两个 /ai/ 双元音的实现：[ɐe] 和 [ʌi] 你能识别包含这些的词吗？

(f) SSE 说话者据说有单元音 /e/ 和 /o/，与 RP 双元音 /eɪ/ 和 /oʊ/ 不同。这对这位演讲者来说是真的吗？

(g) 据说说 SSE 的人在所有音节位置都有 'dark l' ([ɫ])。这对这位演讲者来说是真的吗？

(h) 说话者既有逼近音 [ɹ] 又有拍音 [ɾ]。识别包含这些的单词。

4 西德克萨斯英语（拉伯克）

gueqrecard_002

听曲目 13.7。这也是 PAC 录音；演讲者正在朗读以下书面段落：

前几天我在新闻上看到了霍伊特。我不能确定我是否弄对了事实。但我认为他希望拉伯克的农民为新作物种植大豆、黑麦，也许还有水稻。我昨晚收听了农场报道。他们说棉花价格将继续上涨。没有因为下雨而腐烂肯定是件好事。我告诉你，朝露也没多大用处。我仍然认为我们今年会收回所有投资。收割今年的棉花作物是一个挑战。我让我最大的儿子和我们一起工作，因为他已经不在学校了。他喜欢在农场工作。但我觉得他应该从事石油、养牛等行业，或者与农业无关。他刚满十九岁。你知道，他不玩锡兵或他的弓箭之类的玩具。他几乎是个男人，而且是个相当不错的人。我仍然喜欢讲关于我儿子的这个故事。这个故事真实地展示了他如何扭转一切。我的儿子罗伊当时大约十岁。他的朋友汤姆过来玩了。他们决定偷走我邻居的沙滩车，然后开着它兜风。他们回到这片棉田后面，走上一条马道。好吧，那辆沙滩车装满了，先是一个孩子，然后是另一个。汤姆在开车装置。当他们从铁路或其他地方撞到一个大黄铜钉时，他们并没有开太远。轮胎没气了，他们撞到了一根围栏，然后当马车坠毁时，他们全都头朝下掉进灰烬和干泥里。那些孩子根本不知道他们在做什么。之后，我的邻居肯不得不亏本卖掉了这辆沙滩车。我知道我没吃晚饭就让罗伊上床睡觉，这是对他最轻的惩罚。我告诉我儿子他有责任偿还损失。他整个夏天都在肯的马具房里辛勤劳作，这样他就可以对付他。整个炎热的夏天，我儿子都放弃了在社区游泳池游泳。他用现金偿还了肯。罗伊又拉了那样的东西吗？不，他没有。那件事已经过去九年了。他学会了更谨慎地选择他的朋友和他的行为。我从来没有抽过他，也没有打过他，但他学会了遵守规则，不再偷窃或撒谎。法律放过了他，因为他偿还了肯。他不是个坏孩子，我的罗伊，而且他在其他时候也没有做出正确的选择。只有一次，他想有一天他想拿出一根热棒，幸运的是他的错误没有毁了他的生活。Roy 现在已经成为我二儿子 Beau 和小儿子 Luke 的好榜样。

I saw Hoyt on the news the other day. I can’t be sure whether I got the facts right. But I think he wants Lubbock farmers to plant soy, rye, and maybe rice for the new crop. I tuned in to the farm report last night. They say cotton prices will continue to rise. It sure is good that rot didn’t set in from all the rain. I tell you, the morning dew wasn’t much help either. I still think we will get back all our investment this year. Getting in the cotton crop this year was a challenge. I had my oldest boy at work with us since he isn’t in school anymore. He likes to work on the farm. But I feel he should work in a business like oil, cattle ranching, or maybe nothing to do with agriculture. He just turned nineteen. You know, he doesn’t play with toys like tin soldiers or his bow and arrow. He’s almost a man, and quite a fine one at that. There’s this story about my boy that I still do like to tell. It’s a tale that really shows how he turned everything all around. My son Roy was about ten years old at the time. His friend Tom came over to play. They decided to steal my neighbor’s dune buggy and to go for a joy ride in it. They went back behind this cotton field onto a horse trail. Well, that dune buggy filled up, with first one kid and then another. Tom was at the wheel driving this contraption. They didn’t drive too far when they hit a big brass nail, from the railroad or something. The tire went flat, they hit a holding pen and then they all fell head first into ashes and dried mud when the buggy crashed. Those kids didn’t have good sense at all about what they were doing. My neighbor Ken had to sell the dune buggy at a loss after that. I know I sent Roy to bed without dinner, and that was the least of his punishments. I told my son he had a duty to pay back the damages. He toiled all summer in Ken’s tack room so he could do right by him. My boy gave up swimming at the community pool that whole hot summer. He used the cash to pay back Ken. Did Roy pull something like that again? No he did not. It’s been nine years since that happened. He learned to choose his friends and his actions more carefully. I never used a lash or hit him, but he learned to toe the line and not to steal or lie ever again. And the law left him alone because he paid Ken back. He’s not a bad boy, my Roy, and he didn’t fail to make good choices other times. Just once he thought he’d like to take out a hot rod one day and got lucky that his mistake didn’t ruin his life. Roy has now become a good role model for Beau, my middle son, and for Luke, the youngest.

(a) 说话者是有皱纹的还是没有皱纹的？引用录音中的例子。

(b) 这个说话人对 /aɪ/ 双元音（在词汇集 PRICE 的词中）有可变的“平滑”（单元音化），导致 [a:] 发音。举例说明可变性。

(d) 你能在元音 /æ/、/ε/ 和 /ɪ/ 中听到任何南方变音（双元音化）的例子吗？

(e) 据说许多德州英语变体在结尾 /n/ 之前表现出（可变地）中和 /ɪ/ 与 /ε/ 的区别。这对这位演讲者来说是真的吗？

5 澳大利亚英语

gueqrecard_002

听曲目 13.8。这是 PAC 录音。演讲者正在阅读与练习 3 中 SSE 演讲者相同的单词列表。

(a) 据说澳大利亚英语是非变态英语。这对这位演讲者来说是真的吗？

(b) 澳大利亚英语据说在所有音节位置都有“dark l”。这对这位演讲者来说是真的吗？

(d) 我们在讨论澳洲英语的时候，说过元音音素系统和RP一样，但是有元音移位现象。录音中是否存在这些内容？

6 印度英语（孟买/孟买）

gueqrecard_002

听曲目 13.9。这是 PAC 录音。请参阅 PAC 网站上的单词列表 2。

(a) 说话者是可变的非风湿性的。举例说明 rhotic 和非 rhotic 发音。

(b) 在录音中引用 TH 停止的例子。

(d) 说话者对 /l/ 音素有“清晰”（颚化）的认识吗？

(e) 说话人是如何意识到 /r/ 音位的？

7 练习1-6中重音与RP和GA的比较

gueqrecard_002

收听曲目 13.10和13.11。这些是 RP 和 GA 发言人的 PAC 录音。他们正在阅读 PAC 项目中的以下段落：

一位电视布道者的圣诞节采访

Christmas interview of a television evangelist

如果电视布道者和我们其他人一样，那么他们在圣诞节那周真正想做的就是对家人大发雷霆，批评他们的朋友，隔着花园围栏瞪着邻居的孩子哭。然而，社会期望他们像一年中其他五十一个星期一样快乐和喜气洋洋。如果有的话，更是如此。

If television evangelists are anything like the rest of us, all they really want to do in Christmas week is snap at their families, criticize their friends and make their neighbours’ children cry by glaring at them over the garden fence. Yet society expects them to be as jovial and beaming as they are for the other fifty-one weeks of the year. If anything, more so.

以牧师彼得·“皮特”·史密斯为例，这位“电视牧师”发布了新闻稿，称自己是“捕捉时代精神的人”。他说，在我们上午 9 点在他位于伦敦南部克劳肖大道的“媒体办公室”会面之前，他面临着一个真正的困境。他应该努力“表现得像个基督徒”——打开他的门，给我倒杯茶——还是他应该保持冷静，以一种表现出无聊冷漠的方式研究他的指甲，然后尽快摆脱我尽可能？最后，他两者都没有。

Take the Reverend Peter ‘Pete’ Smith, the ‘TV vicar’ who sends out press releases in which he describes himself as ‘the man who has captured the spirit of the age’. Before our 9 a.m. meeting at his ‘media office’ on Crawshaw Avenue, South London, he faced, he says, a real dilemma. Should he make an effort ‘to behave like a Christian’ – throw his door open, offer me a cup of tea – or should he just play it cool, study his fingernails in a manner that showed bored indifference and get rid of me as quickly as possible? In the end, he did neither.

“事实上，约翰，”他用一种响亮的河口英语口音说，“圣弗朗西斯说，‘任何时候都要传福音，必要时就说话。’” ’但是，嘿，他并不是说“表现出你最好的行为，一直开心。” 我本可以对你格外礼貌，但真正的我会在我说话的时候显露出来。你无法掩饰你的本性。

‘As a matter of fact, John’, he says in a loud Estuary English twang, ‘St Francis said, “At all times preach the gospel and speak whenever you have to.” ’ But hey, he didn’t mean “Be on your best behaviour and be happy all the time.” I could have been extra-polite to you, but the real me would have come out as I was talking. You cannot disguise what you are.’

“那你是什么人，皮特？”

‘And what are you then, Pete?’

“嗯，我是基督徒，约翰。我从 14 岁起就是一个人。而且我确信，人们会更多地根据你是谁而不是你必须说的来评判基督教。许多教会领袖似乎并不理解这一点。他们认为只有当我们将创造的教义强加于人们的喉咙时，我们才能成为真正的基督徒。但如果你试图强行给人们喂食，他们就会厌倦并认为你很痛苦。戴上狗项圈，穿紫色衣服，并且总是谈论新约的真正含义，这被视为基督教领袖的工作。事实上，这让人们马上就厌烦了！

‘Well, I’m a Christian, John. I’ve been one since I was 14. And I know for sure that Christianity will be judged more on who you are rather than what you have to say about it. Many church leaders don’t appear to understand this. They think we can only be really Christian when we are ramming the doctrine of the Creation down people’s throats. But if you try to force-feed people they get sick of it and think you’re a pain. It’s seen as the job of a Christian leader to wear a dog-collar and dress in purple and always be talking about the real meaning of the New Testament. In reality, that turns people right off!’

在许多方面，“皮特”史密斯看起来就像你期望的高调、重生的基督徒的样子：高大、英俊、轮廓分明、晒得均匀。他的牙齿白得吓人，电视布道者往往都有，而且他没有戴狗项圈。事实上，在星期天早上做各种宗教节目时，众所周知，他会穿一件休闲风格的黑色皮夹克。如今，他的造型更加商务：金属框眼镜、灰色西装、蓝色开领衬衫和时尚的黑色大搭扣鞋。史密斯今年 44 岁，但他看起来只有 24 岁。

In many ways, ‘Pete’ Smith looks exactly how you’d expect a high-profile, born-again Christian to look: tall, handsome, clean-cut and evenly sun-tanned. He has those scarily white teeth that TV evangelists tend to have, and he doesn’t wear a dogcollar. In fact, when doing his various religious programmes on Sunday mornings, he has been known to wear a black leather jacket instead, in casual mode. Today, the look is more business-like: metal-rimmed glasses, a grey suit, a blue open-neck shirt, and fashionable black shoes with large buckles. Smith is 44 but he looks a mere 24.

在整个采访过程中，没有谈到穷人或有需要的人，只谈到他即将在二月份的中国之行和在那里等待他的消息的群众。我冒昧地问了几个关于他十年前创立的慈善信托基金的问题，人们普遍认为，该信托基金拥有 800 名员工，在世界各地开办学校、医院和旅馆。而他一直愿意出谋划策的赌博组织呢？这是一项临时活动，还是他已经接受了担任董事会成员的报酬？这些天宗教是哪一边？钱重要吗？就好像我朝他的方向发射了几枚导弹。他只是叹了口气回答：“我只是个人，约翰。上帝知道我尽了最大努力，但经常失败，但如果我们的敌人嘲笑我们所做的一些好事，那我也无可奈何。

During the whole interview, there wasn’t any talk of the poor or the needy but only of his forthcoming trip to China in February and the masses waiting for his message there. I ventured a few questions relating to the charity trust he founded some ten years ago and which, it is generally agreed, employs eight hundred staff and runs schools, hospitals and hostels around the world. And what about the gambling organization he has been willing to advise? Is that a temporary activity or might it be true that he has accepted to be paid to sit on its Board of Directors? Which side is religion on these days? Does money matter? It was as if I had launched a few missiles in his direction. He just sighed in answer: ‘I’m only human, John. God knows I do my best and often fail, But it’s no skin off my nose if our enemies sneer at some of the good work we do. Truth will out.’

(a) 听练习 1-7 中所有口音中 /r/ 音素和 /l/ 音素的实现。你发现什么不同？

(b) 听词汇集 LOT 中的元音（例如not, hot, cot）。你发现什么不同？许多说美国的人在词汇集 LOT 的单词中缺少圆唇元音 [ɒ]，例如possible和dog。相反，据说它们有一个不圆唇的 [ɑ] 元音，因此像pocket这样的词对英国听众来说听起来像packet 。GA 扬声器是否如此？

(d) 听在 RP 中有词尾 schwa 的词（例如leader, Cinderella）。在不同的说话者中，你发现这些词中的元音是什么？

(e) 在非流音重音中，据说 RP 有居中双元音 /ɪə/、/εə/ 和 /ʊə/，在词汇集 NEAR、SQUARE 和 CURE 中。您在非震颤说话者中发现了这些音素的哪些实现？

参考

References

对 IViE 和 PAC 项目的引用是：

The references to the IViE and PAC projects are:

Carr, P.、Durand, J. 和 Pukli, M. (2004)。“PAC 项目：原则和方法”。La Tribune Internationale des Langues Vivantes 36：24-35。本文的 pdf 版本可在 PAC 网站的“出版物”下找到：http: //www.projet-pac.net/

Carr, P., Durand, J. and Pukli, M. (2004). ‘The PAC project: principles and methods’. La Tribune Internationale des Langues Vivantes 36: 24–35. A pdf version of this article can be found under ‘publications’ on the PAC website: http://www.projet-pac.net/

Grabe, E.、Nolan, F. 和 Farrar, K. (1998)。“IViE：英语语调变化的比较转录系统”。1998 年口语处理国际会议论文集。悉尼，澳大利亚。

Grabe, E., Nolan, F. and Farrar, K. (1998). ‘IViE: a comparative transcription system for intonational variation in English’. Proceedings of the 1998 International Conference for Spoken Language Processing. Sydney, Australia.

有关 PAC 项目的更多信息，请参阅：

For more on the PAC project, see:

Przewozny, A.、Moore, S. 和 Turcsan, G.（编辑）（2012 年）。PAC 项目的完整指南。Carnets de Grammaire , CLLE-ERSS, Université Toulouse II-Le Mirail。

Przewozny, A., Moore, S. and Turcsan, G. (eds.) (2012). The Complete Guide to the PAC Project. Carnets de Grammaire, CLLE-ERSS, Université Toulouse II-Le Mirail.